mums»:

 

UBRARY
206C! Michigan State

“:1 ~‘~--\.L.
'1 I"! albily

 

This is to certify that the
thesis entitled

ANALYSIS OF COLD ACCLIMATION ABILITY AND
DROUGHT TOLERANCE OF PETUNIA SPP.

presented by

AARON EMERY WALWORTH

has been accepted towards fulfillment
of the requirements for the

Master of degree in Plant Breeding and Genetics -
Science Horticulture

 

 

 

7 Major Professor’s Signature
I / il / 0‘7

Date

MSU is an Affirmative Action/Equal Opportunity Employer

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K'IProlecc&Pres/CIRC/DateDuetindd

 

ANALYSIS OF COLD ACCLIMATION ABILITY AND DROUGHT
TOLERANCE OF PETUNIA SPP.
By

Aaron Emery Walworth

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE
Plant Breeding and Genetics - Horticulture

2009

ABSTRACT

ANALYSIS OF COLD ACCLIMATION ABILITY AND DROUGHT TOLERANCE
OF PET UNIA SPP.

By
Aaron Emery Walworth

Freezing tolerance is a dynamic characteristic, and many plant species increase in
freezing tolerance following exposure to low non-freezing temperatures, a process
referred to as cold acclimation. The genus Petunia is composed of diverse species, with
the capacity for cold acclimation already known to exist in at least one species, P.
hybrida. In this study, P. exserta, P. integrifolia, and two accessions of P. axillaris were
also found to cold acclimate. All Petunia species had similar basal freezing tolerance of
ELso = -2 °C, but freezing tolerance varied significantly among species following cold
acclimation. Petunia axillaris (accession 28548) showed the greatest acclimated freezing
tolerance with an EL50 temperature of -8 °C, compared to only -5 °C for P. exserta.
Temperature, but not photoperiod, was critical for induction of cold acclimation in P.
hybrida. Cold acclimation of Arabidopsis is largely controlled by genetic factors in the
CBF cold-response pathway. High levels of constitutive heterologous AtCBF 3
expression in P. hybrida ‘Mitchell’ resulted in increased basal freezing tolerance of one
transgenic line by ca. 2.5 °C, while expression of LeCBFI (a CBF homolog from tomato)
had no effect on freezing tolerance. However, high expression of AtCBF 3 resulted in
phenotypic changes including delayed flowering. Heterologous CBF expression did not
enhance drought tolerance. Expression of putative endogenous CBF transcription factors,
petCBF1-4, was induced at 3 °C and two putative downstream genes of the petunia CBF

pathway whose expression was induced by cold and CBF overexpression were identified.

To my wonderful wife, Nicole, for all of her love and support as I worked on this
research and for postponing her dreams as I pursued mine.

iii

ACKNOWLEDGMENTS

I would like to express my thanks to Dr. Ryan Warner, my major professor, for
patiently guiding me through these years of research and allowing me to work in his lab.

Thank you to my guidance committee members, Dr. Bert Cregg and Dr. Michael
Thomashow for helping me along the way, and whose thoughtful input was certainly an
invaluable resource.

Thanks also to Mike Olrich and all of the undergraduate student employees who
helped me in the greenhouse and laboratory.

I would also like to thank my officemate, Joseph Tychonievich, for listening and
providing many helpful suggestions as I worked through problems.

Thank you to everybody else at MSU who has helped me out in one way or
another as I conducted my research project.

A big thank you also goes to the best parents in the world for constantly
supporting me as I spent these last eight years in college.

Finally, thanks to my terrific wife, Nicole, for her love, support, and

encouragement, and for never letting me give up and get a “real job.”

iv

TABLE OF CONTENTS

LIST OF TABLES ......................................................................................................... vii

LIST OF FIGURES ........................................................................................................ ix

CHAPTER 1

INTRODUCTION AND LITERATURE REVIEW ........................................................ 1
Introduction ........................................................................................................... 1
Literature Review .................................................................................................. 2
References ........................................................................................................... 15

CHAPTER 2

CHARACTERIZTNG COLD ACCLIMATION ABILITY OF PET UNIA SPP ............ 20
Abstract ............................................................................................................... 20
Introduction ......................................................................................................... 21
Materials and Methods ........................................................................................ 26
Results ................................................................................................................. 29
Discussion and Conclusions ............................................................................... 31
References ........................................................................................................... 42

CHAPTER 3

EVIDENCE FOR CONSERVATION OF CBF COLD-RESPONSE PATHWAY AND

ITS ROLE IN COLD ACCLIMATION OF PETUNIA H YBRIDA ‘MITCHELL’ ........ 45
Abstract ............................................................................................................... 45
Introduction ......................................................................................................... 46
Materials and Methods ........................................................................................ 50
Results ................................................................................................................. 60
Discussion and Conclusions ............................................................................... 64
References ........................................................................................................... 85

CHAPTER 4

ASSESSING EFFECTS OF CBF OVER-EXPRESSION ON HORTICULTURAL

TRAITS IN PETUNIA HYBRIDA ‘MITCHELL’ .......................................................... 88
Abstract ............................................................................................................... 88
Introduction ......................................................................................................... 88
Materials and Methods ........................................................................................ 90
Results ................................................................................................................. 91
Discussion and Conclusions ............................................................................... 94
References ........................................................................................................... 99

CHAPTER 5

ASSESSING DROUGHT TOLERANCE OF CBF OVER-EXPRESSING PET UNIA

HYBRIDA ‘MITCHELL’ .............................................................................................. 101
Abstract ............................................................................................................. 101

Introduction ....................................................................................................... 101

Materials and Methods ...................................................................................... 102
Results ............................................................................................................... 104
Discussion and Conclusions ............................................................................. 105
References ......................................................................................................... 109

vi

LIST OF TABLES

”C

Table age

2.1 Growth conditions (temperature, photoperiod, photosynthetic photon flux
(PPF) and duration) of P. hybrida 'Mitchell' plants prior to electrolyte leakage
assays .............................................................................................................................. 36

2.2 ANOVA for effect of acclimation regime on freezing tolerance of P. hybrida
'Mitchell' as measured by electrolyte leakage assays ...................................................... 36

2.3 Pairwise comparisons with Fisher's LSD between ELso values of P. hybrida
‘Mitchell’ grown under different acclimation regimes. Shown are p-values for
each comparison .............................................................................................................. 36

2.4 AN OVA for effect of species on nonacclimated (A) and acclimated (B) EL50
temperature as determined by electrolyte leakage assay. ............................................... 37

2.5 Pairwise comparisons with Fisher's LSD between acclimated EL50 values of
different Petunia species. Shown are p-values for each comparison ............................. 37

3.1 Primer sequences and reaction conditions used for screening putative
transgenic P. hybrida ‘Mitchell’ plants for transgene presence in the To
generation ........................................................................................................................ 70

3.2 Primer sequences and reaction conditions used in RT-PCR analysis of CBF
expression in transgenic P. hybrida 'Mitchell' ................................................................ 70

3.3 Plant generation and transgene copy number for the transgenic P. hybrida
‘Mitchell’ lines tested in freezing tolerance experiments ............................................... 71

3.4 Primer sequences and reaction conditions for RT-PCR of putative
endogenous petunia CBF genes. Reaction conditions consisted of 26 cycles of
94 °C, 30 s; 56 °C, 1 min; 72 °C, 2.5 min; plus final extension of 72 °C, 10 min ........... 71

3.5 Putative orthologs of CBF-regulated genes and their expression in response
to cold and CBF overexpression in Petunia hybrida ...................................................... 72

3.6 Primer sequences and reaction conditions used in RT-PCR analysis of
putative downstream genes in the endogenous CBF-pathway in Petunia hybrida
'Mitchell' .......................................................................................................................... 73

3.7 ANOVA for effect of transgenic line on non-acclimated (A) and acclimated
(B) EL50 temperature of AtCBF 3-expressing P. hybrida ‘Mitchell’ lines ...................... 73

vii

3.8 Pairwise comparisons with Fisher's LSD between non-acclimated ELso
values of different AtCBF3-expressing lines and controls. Shown are p-values
for each comparison ........................................................................................................ 74

3.9 AN OVA for effect of transgenic line on non-acclimated (A) and acclimated
(B) ELso temperature of LeCBFI-expressing P. hybrida ‘Mitchell’ lines ...................... 74

4.1 Comparison of horticultural traits of transgenic lines with wild type Petunia
hybrida 'Mitchell' in first replication. All measurements were taken the day the
first flower was fully open. ............................................................................................. 97

4.2 Comparison of horticultural traits of transgenic and wild type Petunia
hybrida 'Mitchell' in second replication. All measurements were taken the day
the first flower was fully open ........................................................................................ 97

5.1 Survival rates of transgenic P. hybrida 'Mitchell' plants expressing AtCBF 3
(pMPSl3 and BpMPSl3) or LeCBFI (pXINl and BpXIN 1) following 9 or 12

days of water withholding and 7 days of recovery. Numbers shown represent

percent survival of 12 individuals. ................................................................................ 107

5.2 Relative gain of above-ground dry biomass averaged for 12 plants per

genotype in each drought treatment. Relative gain of drought plants was defined

as the percent of non-stressed weight gain achieved by stressed plants; calculated

as: (droughtfinal - initial) / (nonstressedfinal - initial) X 100 .................................... 107

5.3 ANOVA for effect of transgenic line on relative biomass gain for 9-day
drought period ............................................................................................................... 108

5.4 ANOVA for effect of transgenic line on relative biomass gain for 12-day
drought period ............................................................................................................... 108

viii

LIST OF FIGURES

"D

Figure age

2.1 Percentage of electrolyte leakage measured at various freezing temperatures
on leaf discs of P. hybrida ‘Mitchell’ grown under different conditions. (A)
nonacclimation under long days (NONLD), (B) nonacclimation under short days
(NONSD), (C) cold acclimation regime 1 (CA1), (D) cold acclimation regime 2
(CA2), (E) cold acclimation regime 3 (CA3), (F) rampdown under short days
(Rampdown SD), (G) rampdown under long days (Rampdown LD). NONLD:
22°C, LD (16hr days). NONSD: 22°C, LD; then 3 weeks at 22°C, SD (9hr days).
CA1: 22°C, LD; then 1 week at 3°C, SD. CA2: 22°C, LD; then 2 weeks at 22°C,
SD; then 1 week at 3°C, SD. CA3: 22°C, LD; then 3 weeks at 3°C, SD.
RampdownSD: 22°C, LD; then 1 week at 15°C, SD; then 1 week at 10°C, SD;
then 1 week at 3°C, SD. Rampdown LD: 22°C, LD; then 1 week at 15°C, LD;
then 1 week at 10°C, LD; then 1 week at 3°C, LD. Error bars represent standard
deviation of six measurements ........................................................................................ 38

'LL’iJ‘l‘ ‘

 

. l‘l,’

2.2 EL50 values for P. hybrida ‘Mitchell’ after various cold acclimation regimes.
NONLD: 22°C, LD (16hr long days). NONSD: 22°C, LD; then 3 weeks at 22°C,

SD (9hr short days). CA1: 22°C, LD; then 1 week at 3°C, SD. CA2: 22°C, LD;

then 2 weeks at 22°C, SD; then 1 week at 3°C, SD. CA3: 22°C, LD; then 3

weeks at 3°C, SD. RampdownSD: 22°C, LD; then 1 week at 15°C, SD; then 1

week at 10°C, SD; then 1 week at 3°C, SD. RampdownLD: 22°C, LD; then 1

week at 15°C, LD; then 1 week at 10°C, LD; then 1 week at 3°C, LD. Bars with

the same letter are not statistically different according Fisher’s LSD with a =

0.05. Error bars represent standard deviation of the two ELSO values calculated

from two replications for each treatment ........................................................................ 39

2.3 Average percent electrolyte leakage at each temperature tested for Petunia

hybrida (A) and wild Petunia species (B-E). Nonacclimated plants were grown

at 22°C LD and acclimation was accomplished by exposing plants to 15°C SD

for 1 week, 10°C SD for 1 week, then 3°C SD for 1 week. Leakage data at each
temperature point is averaged over n measurements. Error bars indicate standard
deviation .......................................................................................................................... 40

2.4 EL50 temperatures for nonacclimated (7 weeks at 22°C LD) and acclimated (6
weeks at 22°C LD, 1 week at 15°C SD, 1 week at 10°C SD, and 1 week at 3°C

SD) wild Petunia species. Bars depict ELso temperatures averaged over at least

2 replications for each species. Bars with the same letter are not statistically

different according Fisher’s LSD with a = 0.05. Error bars indicate standard

deviation .......................................................................................................................... 41

ix

3.1 Diagram of constructs used for Agrobacterium-mediated transformation of P.
hybrida ‘Mitchell.’ All constructs contain NPTII as a selectable marker for

kanamycin resistance. pSPUD73 contains Arabidopsis thaliana CBF] behind

the cold-inducible AtCor 78 promoter. pSPUD74 contains Arabidopsis thaliana

CBF] behind the cold-inducible AtC0r15a promoter. pMPS l 3 and pXIN 1

contain Arabidopsis thaliana CBF 3 and Lycopersicon esculentum CBF] ,

respectively, behind the strong constitutive CaMV 35S promoter .................................. 75

3.2 Gene expression analysis by semi-quantitative RT-PCR of P. hybrida
‘Mitchell’ transgenic lines containing 358: :AtCBF 3 (A) or 35S::LeCBF1 (B).
RNA was isolated from plants grown at 22 °C ............................................................... 75

3.3 RT-PCR analysis of AtCBF 1 expression in pSPUD74 transgenic lines
following various exposures to cold temperatures. Nonacclimated plants were

grown at 22 °C and rampdown acclimated plants were grown 7 d at 15 °C SD, 7
dath°CSD,and7dat3°CSD ................................................................................... 76

3.4 Percent electrolyte leakage at each temperature tested for wild type (A),

empty vector control (B), and transgenic lines (C-E) containing the

35S: :AtCBF 3 (pMPSl3) construct. Nonacclimated plants (black bars) were

grown at 22 °C LD and acclimation was accomplished by exposing plants to 15

°C SD for 1 week, 10 °C SD for 1 week, then 3 °C SD for 1 week (grey bars).

Leakage data at each temperature averaged over 6 measurements for empty

vector and transgenic lines; 24 measurements for wild type. Standard deviation

shown by error bars ......................................................................................................... 77

3.5 ELso temperatures of nonacclimated and acclimated AtCBF 3 constitutively
over-expressing lines. pMPS 1 3-7 is significantly more freezing tolerant than the
control lines prior to acclimation (starred bar). Following our acclimation

regime (7 d at 15 °C SD, 7 d at 10 °C SD, and 7 d at 3 °C SD), there is no

significant difference between any of the transgenic lines and the control lines.

Standard deviation shown by error bars .......................................................................... 78

3.6 Percent electrolyte leakage at each temperature tested for wild type (A),

empty vector control (B), and transgenic lines (C-F) containing the

35S::LeCBF1 (pXIN 1) construct. Nonacclimated plants (black bars) were grown

at 22 °C LD and acclimation was accomplished by exposing plants to 15 °C SD

for 1 week, 10 °C SD for 1 week, then 3 °C SD for 1 week (grey bars). Leakage

data at each temperature averaged over 6 measurements for empty vector and
transgenic lines; 24 measurements for wild type. Standard deviation shown by

error bars ......................................................................................................................... 79

3.7 EL50 temperatures of nonacclimated and acclimated LeCBFI constitutively
over-expressing lines. Nonacclimated plants were grown at 22 °C LD and

acclimation was accomplished by exposing plants to 15 °C SD for 1 week, 10 °C

SD for 1 week, then 3 °C SD for 1 week. There are no significant differences

between any of the transgenic lines and the control lines ............................................... 80

3.8 Alignment of petCBF amino acid sequences obtained from Goldman et al.
(2007) and AtCBFI sequence (Pubmed Gene ID: 828653). Shown in boxes is
the matching of the petCBF sequences with the “CBF signature sequences”,

PKK/RPAGRXKFXETRHP and DSAWR (Jaglo et al. 2001). Grey shading
denotes where the petCBF amino acids differ from the signature sequence .................. 81

3.9 RT-PCR analysis for expression of petCBF I -4 in wild type P. hybrida

‘Mitchell’ and pSPUD74 transgenic lines following chilling at 3 °C for various

time periods. Nonacclimated plants were grown at 22 °C and rampdown

acclimated plants were grown at 15 °C SD for 1 week, 10 °C SD for 1 week, then

3 °C SD for 1 week. WT3-2 is a wild type line recovered from tissue culture and

P. hybrida ‘Mitchell’ is a wild type that has not undergone tissue culture .................... 82

3.10 Southern hybridization with genomic DNA from four Petunia species

digested with PST I (A) or ECO RI (B) restriction enzymes. Probe is a 163

nucleotide fragment from a highly conserved region of petCBF I from P. hybrida
‘Mitchell’ ........................................................................................................................ 83

3.11 Expression of putative downstream components of the CBF-regulon in P.

hybrida ‘Mitchell’ determined by RT-PCR. Cold-responsiveness of sequences

was verified in wild type plants (A) and CBF-responsiveness was determined in
nonacclimated transgenic lines (B). TC numbers and sequences were obtained

from http://compbio.dfci.harvard.edu/tgi/ ....................................................................... 84

3.12 Phylogenetic tree showing relationship between nucleic acid sequences for
CBF transcription factors from various species .............................................................. 84

4.1 Photographs comparing phenotype of transgenic lines and wild type on the day the

first flower opened. Shown are representative individuals from the transgenic lines with
the highest level of heterologous CBF expression and wild type. Images are sized to the
same approximate scale .................................................................................................. 98

xi

CHAPTER 1:

INTRODUCTION AND LITERATURE REVIEW

INTRODUCTION

F loriculture in the United States is a major industry worth over $4.1 billion in
wholesale value in 2007. Within the floriculture industry, bedding/ garden plants make up
the largest portion of sales at over $1.76 billion (USDA-NASS 2008). The majority of
bedding plant sales occur during a short period in the spring and early summer months.
However, a few of the more cold tolerant species such as pansy and ornamental kale have
the ability to withstand multiple frosts, allowing them to be sold earlier in the spring and
later into the fall. Consumers wishing to add color to their spring and late fall landscapes
have traditionally been limited to these and a few other species because few herbaceous
plants tolerate the environmental conditions at this time of year. Additionally,
greenhouse growers have been very dependent upon springtime sales to provide the
majority of their annual income (Kessler 2004). Increasing the cold tolerance of other
bedding plants would add to the number of species offered for sale in the cooler months.
Developing additional cold tolerant cultivars of already-popular bedding plants would
also allow growers to keep their greenhouses productive during summer months by
producing plants for sale in the fall.

Garden petunias (Petunia hybrida) are already a very popular bedding plant,
ranking first in sales among bedding plants in 2007, with a wholesale value of over $111

million (USDA-NASS 2008). The current popularity of petunias can be partially

attributed to recent breeding efforts which have combined novel characteristics, such as
prostrate growth habits, with increased environmental stress tolerances (Griesbach 2007).
Petunia is therefore a prime target for efforts aimed at further increasing the cold and
drought tolerance of bedding plants.

Understanding the genetics controlling abiotic stress tolerance traits is critical for
trait improvement through biotechnology. Biotechnological approaches, such as
transgenic manipulation of gene expression, require identification of genes which
increase stress tolerance when up- or down-regulated. Traditional breeding methods
require the identification of germplasm with elite qualities that can be used for crosses.
This present study adds to our understanding of cold and drought tolerance traits in
Petunia spp. and can serve as a starting point for future efforts to create more stress

tolerant cultivars.

LITERATURE REVIEW

The genus Petunia

There is disagreement among taxonomists about how many species belong in the
genus Petunia, but numbers generally range from 11 (Kulcheski et al. 2006; Lorenze-
Lemke 2006) to 16 species (Griesbach 2007). The genus Petunia can be further divided
into two subgenera, Pseudonicotiana and Eupetunia. Plants of Pseudonicotiana have a
salver-shaped corolla with filaments attached to the middle of the corolla tube. Plants in
Eupetunia have a funnel-shaped corolla and filaments which attach below the middle of
the corolla tube (Griesbach 2007). Regardless of the number of species in the genus, all

species originate in South America (Kulcheski 2006). The southeastern Sierra region of

Brazil has been identified as one of the centers of diversity (Lorenz-Lemke 2006). All
Petunia species have 2n=14 chromosomes; the feature distinguishing them from the
closely related genus, Calibrachoa, which has 2n=l 8 (Griesbach 2007). While origin
and chromosome numbers are similar for all Petunia species, there is considerable
diversity within the genus in terms of flower color, morphology, and pollination
syndrome. The three wild species chosen for these studies, P. axillaris (Lamarck)
Britton, Stems & Poggenburg, P. integrifolia (Hooker) Schinz & Thellung, and P.
exserta Stehmann, span the range of variation for these traits within the genus.

Petunia axillaris is a white flowered species in the subgenus Pseudonicotiana
(Griesbach 2007). The flowers are strongly fragrant, especially at night, and are
pollinated by the nocturnal hawkmoth (Manduca spp.) (Ando et al. 2001). This species
can be further divided into three subspecies, P. a. ssp. axillaris, P. a. ssp. parodii (Steere)
Cabrera, and P. a. spp. subandina Ando (Ando et al. 2001 ).

Petunia exserta also belongs to the subgenus Pseudonicotiana and possesses
several characteristics unique among other wild petunias. Petunia exserta is the only
omithologically pollinated petunia and is also the only red flowered species within the
genus (Griesbach et al. 1999). Petunia exserta is very rare, being found growing in
shady cracks on only four sandstone towers in the Southeastern Sierra region of Brazil
(Lorenz-Lemke 2006). The red color of P. exserta is the result of a mixture of several
anthocyanin pigments which are distinct from those present in modern red cultivars of P.
hybrida (Ando et al. 2000). It has been speculated that the genetic differences between
P. exserta and P. hybrida will be useful in the breeding of new red-flowered cultivars

(Griesbach et a1. 1999; Ando et al. 2000).

Petunia integrifolia is a scentless, purple flowered species in the subgenus
Eupetunia, that is pollinated by bees (Ando et al. 2001). This species exhibits a high
degree of self-incompatibility (Ando et al. 2001; personal observation), making visits
from pollinating bees very important for reproduction. Similar to P. axillaris, P.
integrifolia can be further divided into several subspecies (Griesbach 2007). Petunia
integrifolia seems to be gaining popularity among home gardeners and is often marketed
as “wild petunia.”

Petunia hybrida ‘Mitchell’ is a doubled-haploid hybrid petunia. This particular
hybrid arose from the anther culture of a plant resulting from the crossing of P. axillaris x
(P. axillaris x P. hybrida ‘Rose du Ciel’). This plant is particularly useful in genetic
studies because the doubled-haploid nature of the cultivar results in homozygosity at all

alleles (Griesbach 2007).

Cold tolerance of plants

Cold temperatures are among the many abiotic stresses that plants must endure.
Plant species can be grouped according to their ability to tolerate cold temperatures. The
least cold tolerant species are referred to as chilling sensitive and suffer damage when
temperatures are cool but remain above freezing. Freezing sensitive species tolerate cool
temperatures, but are damaged when temperatures fall below freezing. The hardiest
species are freezing tolerant and able to withstand temperatures below freezing (Chen and
Li1980)

Cold tolerance is a dynamic characteristic, changing as environmental conditions

change. When grown in warm temperatures, even cold hardy species have low tolerance

to freezing. Exposure to low non-freezing temperatures brings about an increase in cold
tolerance by a process called cold acclimation. This acclimation process allows plants to
survive future temperatures that are much lower than would be tolerated without

acclimation (Thomashow 1999).

Genetics of cold tolerance

Gene expression (Guy et a1. 1985) and cellular metabolite profile (Cook et al.
2004) changes occur during cold acclimation. Expression of hundreds of genes is altered
following exposure to low temperatures. Vogel et al. (2005) defined a set of low
temperature-responsive genes, termed the COS @ld Standard) set, that are reliably up-
or down-regulated in response to low temperature whether plants are grown in soil or on
agar plates. This COS set includes more than 300 genes that are up-regulated in
Arabidopsis in response to low temperature and 212 genes that are downregulated.

The COS set includes members of the previously identified COR (CQld-
Responsive) family of cold-induced genes in Arabidopsis (Baker et al. 1994; Thomashow
1999). These COR genes are regulated by both an ABA-dependent and an ABA-
independent pathway (Gilmour and Thomashow 1991). COR genes such as COR 1 5a
encode a variety of polypeptides which act to increase freezing tolerance (Steponkus et
al. 1998; Thomashow 1999). Other cold-regulated genes encode antifreeze proteins,
signal transduction proteins, and transcription factors (Maruyama et al. 2004, Vogel et al.
2005).

The promoter regions of the cold-induced COR genes contain a eis-acting element

called the Q-repeat/thydration responsive element (CRT/DRE) (Gilmour et al. 1998)

that is necessary for maintaining the cold-responsiveness of the genes (Baker et al. 1994).
This CCGAC sequence is also present in the promoter region of BM 15, a cold-regulated
gene in Brassica napus (J iang et al. 1996), as well as some cold-regulated tomato genes
(Zhang et al. 2004). The cold-induced transcriptional activator CBF] (C-
repeat/Dehydration responsive element binding factor 1) binds to the CRT/DRE sequence
to induce transcription of other cold-responsive genes in Arabidopsis. This
transcriptional activator protein has a molecular mass of 24kDa and contains a nuclear
localization sequence, an acid activation domain, and an AP2 DNA binding domain of 60
amino acids (Stockinger et al. 1997).

CBF], along with two other genes, CBF2 and CBF 3 , comprise a small family of
transcriptional activators in Arabidopsis (Gilmour et al. 2004). All three proteins are
88% identical at the amino acid level and are located in tandem array on chromosome IV.
As with CBFl , both CBF 2 and CBF 3 bind to CRT/DRE sequences in the promoter
regions of cold regulated genes (Gilmour et al. 1998; Medina et al. 1999) and all three
are functionally redundant (Gilmour et al. 2004). Simultaneous to the discovery of the
CBFI-3 genes by the Thomashow lab, the same genes were discovered by the Shinozaki
and Yamaguchi-Shinozaki lab which named them DREBlb, DREBIc, and DREBI a,
respectively (Liu et al. 1998).

The level of CBF transcripts increases within 15 minutes of exposing the plant to
25°C. These levels continue to increase for two more hours and then begin to slowly
fall. Transcript levels remain higher than in non-chilled plants for at least 24 hours.

After two hours of cold exposure, transcripts for cold induced genes COR15a and COR 78

also begin to accumulate (Gilmour et al. 1998).

CBF transcript accumulation is not controlled by a simple on/off mechanism.
Transcript levels do increase in response to a sudden cold shock, but they also increase in
response to gradual cooling (Zarka et al. 2003). Colder temperatures result in greater
transcript accumulation and slightly warmer temperatures result in less transcript
accumulation. The temperature sensing mechanism can become desensitized after a long
period of exposure to a given cold temperature, at which point, CBF expression
decreases. The cold sensing mechanism becomes resensitized after exposure to warm
temperatures for a period of time ranging from 8 to 24 hours. While in the desensitized
state, CBF expression will continue to increase if the temperature is decreased even
further (Zarka et a1. 2003).

Constitutive expression of CBF] in Arabidopsis induces COR gene expression
and increases freezing tolerance of nonacclimated plants (Jaglo-Ottosen et al. 1998).
Likewise, transgenic Arabidopsis overexpressing CBF 3 are more tolerant of salt, drought,
and cold stresses compared to control plants (Kasuga et al. 1999). Negative phenotypic
effects have been observed when CBF 3 is expressed in Arabidopsis behind the strong
constitutive cauliflower mosaic virus (CaMV) 35S promoter. Transgenic plants are
delayed in flowering, smaller in overall size, more prostrate in growth form, and have
shorter petioles (Kasuga et al. 1999; Gilmour et al. 2000). The use of the stress-inducible
rd29A (Car 78) promoter to drive expression of CBF 3 minimizes these negative effects
while still increasing stress tolerance of transgenic plants following cold acclimation
(Kasuga et al. 1999).

Many biochemical changes occur in the CBF 3 overexpressing plants which are

similar to the changes that occur when wild type plants are cold acclimated. For

example, overexpression of CBF 3 results in increased proline levels in non-acclimated
plants and increased transcript levels for P5CS, an enzyme involved in proline
biosynthesis in plants (Gilmour et a1. 2000).

While all three CBF genes appear to be functionally redundant within their roles
of cold-responsive gene regulation, freezing tolerance, and plant development (Gilmour
et al. 2004), CBF 2 seems to also play a role in the negative regulation of CBF] and
CBF 3 (Novillo et al. 2004). The cbf? mutant has increased tolerance to salt, drought, and
freezing stress. The cbf2 plants are actually more freezing tolerant than wild type plants
with or without acclimation. Basal expression of CBF] and CBF 3 is increased in the
cbf2 mutant and high expression levels are sustained longer in response to cold
temperatures. In wild type plants, upon exposure to stress, CBF] and CBF 3 transcripts
accumulate sooner than CBF2 transcripts. Taken together, these data suggest that CBF 2
is a negative regulator of CBF] and CBF 3 expression (Novillo et al. 2004). ZAT12,
another cold-responsive transcription factor which is induced parallel to CBF1-3, has also
been shown to be involved in the negative regulation of CBF] and CBF 3 (Vogel et al.
2005)

Control of CBF expression is not accomplished through autoregulation since the
promoter regions lack the CCGAC sequence to which CBF binds during the induction of
other cold-regulated genes (Gilmour et al. 1998). CBF expression is regulated by an
upstream transcription factor, ICE] (Inducer of _C_BF Expression), which binds to
promoter regions of the CBF genes. An Arabidopsis mutant defective in ICEl
production, ice], shows reduced CBF 3 expression, lowered COR gene expression when

exposed to cold temperatures, and lower freezing tolerance compared to wild-type plants,

indicating that ICEl positively regulates CBF gene expression. The ICE] gene is
constitutively expressed in the nucleus yet the CBF genes require a cold treatment for
expression. Therefore, a cold-induced modification such as phosphorylation/
dephosphorylation likely occurs to activate the ICEl protein upon exposure to cold
temperature. This change allows ICE] to bind to the promoters of the CBF genes and
induces their expression (Chinnusamy et al. 2003).

As opposed to CBF 3, CBF] and CBF 2 are not as strongly affected by the ice]
mutation, indicating that there may be multiple mechanisms regulating expression of the
three CBF genes (Lee et al. 2005; Chinnusamy et al. 2003). The promoter of CBF2
contains two regions which are sufficient for cold-responsiveness. These two regions,
ICErl and ICEr2 (Induction of CBF Expression regions 1_ and _2_), act together to
stimulate CBF 2 transcription in response to cold and are also involved in gene expression
in response to mechanical agitation. However, it is still unclear which proteins bind to
the ICErl and ICEr2 regions (Zarka et al. 2003).

It has been suggested that CBF proteins need to be activated before they can
function in the induction of cold inducible genes. The sfr6 mutant in Arabidopsis
accumulates normal levels of CBF transcripts following cold exposure but produces low
levels of the cold-induced genes targeted by the CBF proteins, suggesting a role for sfr6
in the activation of CBF proteins (Knight et al. 1999).

The expression of the CBF genes is partially dependent upon the time of day
during which the low temperature stress is present. If the cold stress is present earlier in
the day, then CBF expression will be greater than if the stress is present later in the day.

The highest levels of expression occur when Arabidopsis plants are exposed to low

temperatures 4 hours after dawn and the lowest levels occur at 16 hours after dawn. This
difference is regulated at the transcriptional level (Fowler et al. 2005).

Of the 514 COS genes that have been identified, 93 are in the CBF regulon. Of
these 93 genes in the regulon, 85 are upregulated and 8 are downregulated by low
temperatures. The upregulated genes have a CRT/DRE element where CBF transcription
factors may bind. The downregulated COS genes do not have CRT/DRE elements so
they are most likely not directly targeted by CBF transcription factors (V ogel et al. 2005).
Within the CBF regulon, two other transcription factors, RAP2.1 and RAP2.6, are
activated by CBF in response to cold temperature. These RAP transcription factors may
control subregulons of the CBF regulon (Fowler and Thomashow 2002).

Changes in the metabolome of cold acclimated Arabidopsis plants are largely, but
not entirely, controlled by the action of the CBF genes. In one study, 325 metabolites
were found to increase in response to low temperatures in wild type plants. The
metabolome of nonacclimated CBF 3 overexpressing plants also had increased levels of
256 (79%) of these metabolites. Therefore, a majority of the metabolome profile changes
are under the control of the CBF regulon (Cook et al. 2004).

However, there is evidence for the presence of freezing tolerance genes which are
not controlled by the CBF regulon Warm-grown plants overexpressing CBF 3 with a
constitutive promoter are more freezing tolerant than non-acclimated control plants, but
they become tolerant to even lower temperatures following cold exposure. This provides
evidence for freezing tolerance genes which are not controlled by CBF 3 but which are

activated at low temperatures (Gilmour et al. 2000).

10

The CBF genes are also induced in response to stresses other than cold
temperatures. CBF] and CBF 2 transcripts accumulate in response to mechanical
agitation while CBF 3 does not (Gilmour et al. 1998). CBF1-3 are not responsive to
dehydration stress (Medina et al. 1999) while CBF 4, a fourth member of the CBF family,

is highly induced in response to drought stress (Haake et al. 2002).

Conservation of CBF in other species

CBF appears to be well conserved among the plant species, including monocots
and herbaceous and woody dicots. CBF orthologs have been identified in many species
including Brassica napus L. cv. Jet neuf (Gao et al. 2002), B. napus L. cv. Westar,
Solanum esculentum L., Triticum aestivum L., Secale cereale L. (Jaglo et al. 2001),
Fragaria Xananassa Duchesne, Prunus cerasus L. (Owens et al. 2002), P. avium L.
(Kitashiba et al. 2004), Eucalyptus gunnii Hook. (Kayal et al. 2006), Hordeum vulgare L.
(Choi et al. 2002), F estuca arundinacea Schreb (Tang et al. 2005), and Zea mays L. (Qin
et al. 2004). As with Arabidopsis CBF proteins, the CBF protein orthologs contain the
AP2/EREBP DNA-binding domain and are upregulated in response to cold temperatures.
The orthologs in B. napus, T. aestivum, S. cereale, H. vulgare, F. arundinacea, and E.
gunnii also contain the PKK/RPAGRxKFxETRHP and DSAWR CBF signature
sequences immediately upstream and downstream, respectively, from the AP2/EREBP
domain (Jaglo et al. 2001; Choi et al. 2002; Tang et al. 2005; Kayal et al. 2006).

Ectopic expression of several of these orthologs increases the freezing tolerance
of Arabidopsis. Constitutive expression of ZmDREBIA (a CBF ortholog in maize) in

Arabidopsis induces expression of CBF target genes and results in increased tolerance to

11

drought and freezing stress in non-acclimated plants (Qin et al. 2004). Likewise,
transformation of CIG-B (a CBF ortholog from sweet cherry) into Arabidopsis with a
constitutive promoter induces COR15a expression without stress treatment. These
transgenic plants are more cold and salt tolerant than wild type plants and exhibit stunted
growth (Kitashiba et al. 2004).

Additionally, expressing Arabidopsis CBF genes in several of the previously
mentioned species enhances expression of cold-response genes and increases freezing
tolerance. Constitutive expression of Arabidopsis CBF genes in B. napus cv. Westar
stimulates Bn115 and Bn28 (orthologs of Arabidopsis COR15a and COR6. 6,
respectively) expression in transgenic plants without cold acclimation and increases plant
freezing tolerance (Jaglo et al. 2001). Similarly, the freezing tolerance of nonacclimated
leaves from transgenic strawberry plants, which overexpress AtCBF] under control of the
CaMV 35S promoter, is increased compared to nonacclimated wild type plants.
However, the tolerance of the strawberry plant receptacles to freezing remains unchanged
(Owens et al. 2002).

In the Solanaceae family, freezing-sensitive tomato plants have been shown by
Zhang et al. (2004) to contain three CBF genes (LeCBF1-3). These three CBF genes are
located in tandem array on chromosome III in the tomato genome. All three LeCBF
genes are induced by mechanical agitation but not by ABA, salinity, or drought. LeCBF]
is the only one induced by low temperatures and this induction varies depending on the
photoperiod in which the plant is grown. When grown under constant light, peak levels
of transcript accumulation are reached after 2 hours of cold and return to normal levels

after 24 hours. When plants are grown in a 16 hour photoperiod, transcript levels remain

12

elevated for 16 hours and decrease to slightly above normal levels by 24 hours (Zhang et
al. 2004).

When LeCBF] is overexpressed in Arabidopsis, COR genes are induced and
freezing tolerance is increased. Freezing tolerance does not increase when LeCBF] or
AtCBF] is overexpressed in tomato plants (Zhang et al. 2004), but drought stress
tolerance increases with overexpression of AtCBF 1 (Hsieh et a]. 2002). Only four tomato
genes are induced by the overexpression of either CBF gene. Therefore, tomato plants
do have a CBF regulon, but it is smaller than the CBF regulon present is Arabidopsis
(Zhang et a]. 2004).

Freezing-sensitive Solanum tubersoum (potato) and freezing-tolerant S.
commersonii (wild potato) also respond to CBF overexpression by exhibiting increased
cold tolerance (Pino et a1. 2008). Solanum tuberosum normally has a freezing tolerance
of -3°C before cold acclimation and does not display any increased tolerance following
acclimation. However, when AtCBF] or AtCBF 3 are expressed behind a constitutive
CaMV 35S promoter, freezing tolerance of warm grown plants increases to -5°C (Pino et
al. 2007; Pino et al. 2008). Additionally, expression of these same genes behind a cold-
inducible rd29A promoter does not alter freezing tolerance of warm grown plants, but
increases the freezing tolerance of cold acclimated plants to -5°C following a two-week
treatment at 2°C (Pino et al. 2007).

S. commersonii is a wild relative of the cultivated potato, but unlike S. tuberosum,
this plant is frost tolerant and able to cold acclimate. Prior to acclimation it can survive
temperatures down to -5°C, and following acclimation it can tolerate temperatures as low

as -12°C (Chen and Li 1980). Heterologous expression of AtCBF I with a CaMV 35S

13

promoter in S. commersonii results in an increase in the freezing tolerance of warm
grown plants by 2 to 4 °C. Cold acclimation of these transgenic lines results in further

increases in freezing tolerance by anywhere from 1 to 4 °C (Pino et al. 2008).

Cold acclimation and CBF in petunia

Petunia hybrida is Solanaceous species capable of cold acclimation (Yelenosky
and Guy 1989; Pennycooke et al. 2003). The freezing tolerance of petunia has been
successfully increased by creating transgenic plants with modified expression of u-
Galactosidase (a-Gal) (Pennycooke et a]. 2003).. a-Gal is responsible for breaking down
raffinose oligosaccharides and the activity of a-Gal increases during deacclimation of
petunia resulting in lower raffinose content (Pennycooke et al. 2004). Transgenic
petunias created with reduced u-Gal transcript levels leads to increased raffinose content.
These plants display an increased ability to cold acclimate and are able to increase in
freezing tolerance by 3 to 6°C. On the other hand, plants with increased a-Gal gene
expression have lower raffinose content and lower freezing tolerance (Pennycooke et a1.
2003)

Our present research furthers our understanding of the genetic mechanisms
controlling cold tolerance in Petunia spp. and identifies possible methods for increasing

the cold tolerance of garden petunias.

14

REFERENCES

Ando T., F. Tatsuzawa, N. Saito, M. Takahashi, Y. Tsunashima, H. Numajiri,
H. Watanabe, H. Kokubun, R. Hara, H. Seki, and G. Hashimoto. 2000.
Differences in the floral anthocyanin content of red petunias and Petunia
exserta. Phytochemistry 54:495-501.

Ando T., M. Nomura, J. Tsukahara, H. Watanabe, H. Kokubun, T. Tsukarnoto,
G. Hashimoto, E. Marchesi, and I. Kitching. 2001. Reproductive isolation in a
native population of Petunia sensu Jussieu (Solanaceae). Ann. Bot. 88:403-413.

Baker S., K. Wilhelm, and M. Thomashow. 1994. The 5’-region of Arabidopsis thaliana
cor15a has cis-acting elements that confer cold-, drought- and ABA-regulated
gene expression. Plant Mol. Biol. 24:701-713.

Chinnusamy V., M. Ohta, S. Kanrar, B. Lee, X. Hong, M. Agarwal, and J. Zhu. 2003.
ICEl: a regulator of cold-induced transcriptome and freezing tolerance in
Arabidopsis. Genes Dev. 17: 1043-1054.

Chen H.H. and PH. Li. 1980. Characteristics of cold acclimation and deacclimation in
tuber-bearing Solanum species. Plant Physiol. 65:1 146-1148.

Choi D., E. Rodriguez, and T. Close. 2002. Barley CBF 3 gene identification, expression
pattern, and map location. Plant Physiol. 129: 1 781-1787.

Cook D., S. Fowler, 0. Fiehn, and M. Thomashow. 2004. A prominent role for the CBF
cold response pathway in configuring the low-temperature metabolome of
Arabidopsis. Proc. Natl. Acad. Sci. 101:15243-15248.

Fowler S. and M. Thomashow. 2002. Arabidopsis transcriptome profiling indicates that
multiple regulatory pathways are activated during cold acclimation in addition to
the CBF cold response pathway. Plant Cell 14:1675-1690.

Fowler S., D. Cook, and M. Thomashow. 2005. Low temperature induction of
Arabidopsis CBF 1, 2, and 3 is gated by the circadian clock. Plant Physiol.
137:961-968.

Gao M., G. Allard, L. Byass, A. Flanagan, and J. Singh. 2002. Regulation and
characterization of four CBF transcription factors from Brassica napus. Plant
Mol. Biol. 49:459-471.

Gilmour S. and M. Thomashow. 1991. Cold acclimation and cold-regulated gene
expression in ABA mutants of Arabidopsis thaliana. Plant Mol. Biol. 17: 1233-
1240.

15

Gilmour S., D. Zarka, E. Stockinger, M. Salazar, J. Houghton, and M. Thomashow. 1998.
Low temperature regulation of the Arabidopsis CBF family of AP2 transcriptional
activators as an early step in cold-induced COR gene expression. Plant J. 16:433-
442.

Gilmour S., A. Sebolt, M. Salazar, J. Everard, and M. Thomashow. 2000. Overexpression
of the Arabidopsis CBF 3 transcriptional activator mimics multiple biochemical
changes associated with cold acclimation. Plant Physiol. 124:1854-1865.

Gilmour S., S. Fowler, and M. Thomashow. 2004. Arabidopsis transcriptional activators
CBF], CBF2, and CBF 3 have matching functional activities. Plant Mol. Biol.
54:767-781.

Griesbach R.J., J .R. Stehmann, and F. Meyer. 1999. Anthocyanins in the “red”
flowers of Petunia exserta. Phytochemistry 51:525-528.

Griesbach R]. 2007. Petunia, p.301-336. In: Anderson N.O. (ed.). Flower breeding and
genetics: Issues, challenges and opportunities for the 2lst century. Springer,
Netherlands.

Guy C ., K. Niemi, and R. Brambl. 1985. Altered gene expression during cold acclimation
of spinach. Proc. Natl. Acad. Sci. 82:3673-3677.

Haake V., D. Cook, J. Riechmann, O. Pineda, M. Thomashow, and J. Zhang. 2002.
Transcription factor CBF 4 is a regulator of drought adaptation in Arabidopsis.
Plant Physiol. 130:639-648.

Hsieh T., J. Lee, Y. Charng, and M. Chan. 2002. Tomato plants ectopically expressing
Arabidopsis CBF] show enhanced resistance to water deficit stress. Plant Physiol.
130:618-626.

Jaglo K., S. Kleff, K. Amundsen, X. Zhang, V. Haake, J. Zhang, T. Deits, and M.
Thomashow. 2001. Components of the Arabidopsis C-repeat/dehydration—
responsive element binding factor cold-response pathway are conserved in
Brassica napus and other plant species. Plant Physiol. 127:910-917.

Jaglo-Ottosen K., S. Gilmour, D. Zarka, O. Schabenberger, and M. Thomashow. 1998.
Arabidopsis CBF] overexpression induces COR genes and enhances freezing
tolerance. Science 280: 104-106.

J iang C., B. Iu, and J. Singh. 1996. Requirement of a CCGAC cis-acting element for cold

induction of the BN115 gene from winter Brassica napus. Plant Mol. Biol.
30:679-684.

16

Kasuga M., Q. Liu, S. Miura, K. Yamaguchi-Shinozaki, and K. Shinozaki. 1999.
Improving plant drought, salt, and freezing tolerance by gene transfer of a single
stress-inducible transcription factor. Nature Biotechnology 17:287-291.

Kayal W., M. Navarro, G. Marque, G. Keller, C. Marque, and C. Teulieres. 2006.
Expressed profile of CBF-like transcriptional factor genes from Eucalyptus in
response to cold. J. Expt. Bot. 57:2455-2469.

Kessler J .R. 2004. Growing and marketing bedding plants. Alabama Cooperative
Extension System Bulletin ANR-559

Kitashiba H., T. Ishizaka, K. Isuzugawa, K. Nishimura, and T. Suzuki. 2004. Expression
of a sweet cherry DREBI/CBF ortholog in Arabidopsis confers salt and freezing
tolerance. J. Plant Physiol. 161:1171-1176.

Knight H., E. Veale, G. Warren, and M. Knight. 1999. The sfi6 mutation in Arabidopsis
suppresses low-temperature induction of genes dependent on the CRT/DRE
sequence motif. Plant Cell 11:875-886.

Kulcheski F.R., V.C. Muschner, A.P Lorenz-Lemke, J.R. Stehmann, S.L. Bonatto,
F.M. Salzano, and LB. Freitas. 2006. Molecular phylogenetic analysis of
Petunia Juss. (Solanaceae). Genetics 12623-14.

Lee B., D. Henderson, and J. Zhu. 2005. The Arabidopsis cold-responsive transcriptome
and its regulation by ICE]. Plant Cell 17:3155-3175.

Liu Q., M. Kasuga, Y. Sakuma , H. Abe, S. Miura, K. Yamaguchi-Shinozaki, and K.
Shinozaki. 1998. Two transcription factors, DREBl and DREB2, with an
EREBP/AP2 DNA binding domain separate two cellular signal transduction
pathways in drought- and low-temperature-responsive gene expression,
respectively, in Arabidopsis. Plant Cell 10:1391-1406.

Lorenz-Lemke A.P., G. Mader, V.C. Muschner, J .R. Stehmann, S.L. Bonatto, F.M.
Salzano, and LB. F reitas. 2006. Diversity and natural hybridization in a highly
endemic species of Petunia (Solanaceae): A molecular and ecological
analysis. Mol. Ecol. 15:4487-4497.

Medina J ., M. Bargues, J. Terol, M. Perez-Alonso, and J. Salinas. 1999. The Arabidopsis
CBF genes family is composed of three genes encoding AP2 domain-containing
proteins whose expression is regulated by low temperature but not by abscisic
acid or dehydration. Plant Physiol. 119:463-469.

Maruyama K., Y. Sakuma, M. Kasuga, Y. Ito, M. Seki, H. Goda, Y. Shimada, S.
Yoshida, K. Shinozaki, and K. Yamaguchi-Shinozaki. 2004. Identification of
cold-inducible downstream genes of the Arabidopsis DREBlA/CBF 3
transcriptional factor using two microarray systems. Plant J. 38: 982-993.

17

Novillo F ., J. Alonso, J. Ecker, and J. Salinas. 2004. CBF2/DREB] C is a negative
regulator of CBF I/DREBI B and CBF3/DREBIA expression and plays a central
role in stress tolerance in Arabidopsis. Proc. Natl. Acad. Sci. 101:3985-3990.

Owens C., M. Thomashow, J. Hancock, and A. Iezzoni. 2002. CBF] orthologs in sour
cherry and strawberry and the heterologous expression of CBF] in strawberry. J.
Amer. Soc. Hort. Sci. 127:489-494.

Pennycooke J ., M. Jones, and C. Stushnoff. 2003. Down-regulating u-Galactosidase
enhances freezing tolerance in transgenic petunia. Plant Physiol. 133:901-909.

Pennycooke J ., R. Vepachedu, C. Stushnoff, and M. Jones. 2004. Expression of an (t-
Galactosidase gene in petunia is upregulated during low-temperature
deacclimation. J. Amer. Soc. Hort. Sci. 129:491-496.

Pino M.T., J .S. Skinner, E]. Park, Z. Jeknic, P.M. Hayes, M.F. Thomashow, and T.H.H.
Chen. 2007. Use of a stress inducible promoter to drive ectopic AtCBF expression
improves potato freezing tolerance while minimizing negative effects on tuber
yield. Plant Biotechnol. J. 5:591-604.

Pino M.T., J .S. Skinner, Z. Jeknic, P.M. Hayes, A.H. Soeldner, M.F. Thomashow, and
T.H.H. Chen. 2008. Ectopic AtCBF] over-expression enhances freezing tolerance

and induces cold acclimation-associated physiological modifications in potato.
Plant Cell Environ. 31:393-406.

Qin F ., Y. Sakuma, J. Li, Q. Liu, Y. Li, K. Shinozaki, and K. Yamaguchi-Shinozaki.
2004. Cloning and functional analysis of a novel DREBl/CBF transcription factor
involved in cold-responsive gene expression in Zea mays L. Plant Cell Physiol.

45:1042-1052.

Steponkus P.L., M. Uemura, R.A. Joseph, S. Gilmour, and M. Thomashow. 1998. Mode
of action of the COR15a gene on the freezing tolerance of Arabidopsis thaliana.
Proc. Natl. Acad. Sci. 95:14570-14575.

Stockinger E., S. Gilmour, and M. Thomashow. 1997. Arabidopsis thaliana CBF]
encodes an AP2 domain-containing transcriptional activator that binds to the C-
repeat/DRE, a cis-acting DNA regulatory element that stimulates transcription in
response to low temperature and water deficit. Proc. Natl. Acad. Sci. 94: 1035-

1040.
Tang M., S. Lu, Y. Jing, X. Zhou, J. Sun , and S. Shen. 2005. Isolation and purification of
a cold-inducible gene encoding a putative DRE-binding transcription factor from

F estuca arundinacea. Plant Physiol. Biochem. 43:233-239.

Thomashow M. 1999. Plant cold acclimation: Freezing tolerance genes and regulatory
mechanisms. Annu. Rev. Plant Physiol. Plant Mol. Biol. 50:571-599.

18

United States Department of Agriculture (USDA)-National Agricultural Statistics
Service (NASS). 2008. F loriculture Crops 2007 Summary.
<http://usda.mannlib.comell.edu/usda/current/FlorCrop/FlorCrop-O4-24-
2008.pdf>

Vogel J ., D. Zarka, H. Van Buskirk, S. Fowler, and M. Thomashow. 2005. Roles of the
CBF 2 and ZA T12 transcription factors in configuring the low temperature
transcriptome of Arabidopsis. Plant J. 41:195-21 1.

Yelenosky G. and C. Guy. 1989. Freezing tolerance of citrus, spinach, and petunia leaf
tissue. Plant Physiol. 89:444-451.

Zarka D., J. Vogel, D. Cook, and M. Thomashow. 2003. Cold induction of Arabidopsis
CBF genes involves multiple ICE (Inducer of CBF Expression) promoter
elements and a cold-regulatory circuit that is desensitized by low temperature.

Plant Physiol. 133:910-918.

Zhang X., S. Fowler, H. Cheng, Y. Lou, S. Rhee, E. Stockinger, and M. Thomashow.
2004. F reezing-sensitive tomato has a functional CBF cold response pathway, but
a CBF regulon that differs from that of freezing-tolerant Arabidopsis. Plant J.
39:905-919.

19

CHAPTER 2

CHARACTERIZING COLD ACCLIMATION ABILITY OF PE T UNIA SPP.

ABSTRACT

Freezing tolerance of many plant species increases following exposure to low,
non-freezing temperatures, a process termed cold acclimation. In some species,
shortened photoperiods also bring about an increase in freezing tolerance. Within the
plant family Solanaceae, species vary widely in cold acclimation ability. The objectives
of this work were to examine the roles of low temperature and photoperiod in cold
acclimation of Petunia hybrida and to evaluate cold acclimation of several wild Petunia
species. Temperature, but not photoperiod, influenced cold acclimation of P. hybrida.
Whether grown under long days or short days, nonacclimated plants had an ELSO value
(temperature at which 50% of cellular electrolytes are lost) of ca. —2 °C. Plants
acclimated by gradual cooling at temperatures of 15 °C, 10 °C and 3 °C for 7 days each,
reached an EL50 of ca. -5 °C, regardless of photoperiod. Exposure to 3 °C under short
days for one or three weeks resulted in EL50 temperatures of -3.9 °C and -4.9 °C,
respectively. Freezing tolerance of petunia species P. exserta, P. integrifolia, P. axillaris
(USDA accessions 28546 and 28548), and P. hybrida ‘Mitchell’ was similar prior to cold
acclimation, but varied from -5 °C for P. exserta to -8 °C for P. axillaris (accession

28548) following cold acclimation.

20

INTRODUCTION

Plants endure many biotic and abiotic stresses throughout their lives including
exposure to cold temperatures. Tremendous damage to plant cells can result from
freezing temperatures, primarily to the cellular membranes (Thomashow 1999). Many
plants have adapted to life in cooler climates by developing mechanisms to tolerate low
temperatures, but cold tolerance is a dynamic characteristic. When grown in warm
temperatures, even cold hardy plants have low tolerance to freezing. Exposure to low
non-freezing temperatures, and in some species shortened photoperiods, brings about an
increase in cold tolerance through cold acclimation. Acclimation processes allow plants
to survive temperatures that are much lower than would be tolerated without acclimation
(Thomashow 1999).

Cold acclimation brings about changes in gene expression (Guy et a]. 1985) and
cellular metabolite profiles (Cook et a]. 2004). Of the changes that occur, those regulated
by the CBF cold response pathway are the most thoroughly characterized. In Arabidopsis
thaliana L., this pathway consists of three, functionally redundant transcription factors,
AtCBF1-3 (Gilmour eta]. 2004). Downstream components of the pathway such as the
COR (COM-Responsive) family of genes encode a variety of polypeptides which act to
increase freezing tolerance. For example, COR15a is involved in the stabilization of
membranes during cold temperature exposures (Steponkus et‘ al. 1998; Thomashow
1999). Promoter regions of these downstream genes contain a cis-acting element called
the Q-repear/thydration responsive glement (CRT/DRE) to which CBF binds
(Stockinger et al. 1997; Gilmour et a]. 1998). Expression of the A. thaliana genes

AtCBF1-3 increases in response to low temperatures thereby inducing expression of the

2]

downstream components and ultimately increasing tolerance to freezing temperatures
(Gilmour et al. 1998; Jaglo-Ottosen et al. 1998).

In addition to exposure to low non-freezing temperatures, shortened photoperiods
induce cold acclimation of many woody plant species (Howell and Weiser 1970; Li et al.
2002). Acclimation of Pyrus malus L. occurs in two stages. Shortened photoperiods
induce the first stage and result in an increase in bark hardiness (defined as the highest
temperature at which bark from l-year-old branches is killed) from ~10 °C in
nonacclimated plants down to -25 °C. Frost induces the second stage and results in
acclimation down to -55 °C (Howell and Weiser 1970). In Betula pendula Roth., both
shortened photoperiod (12 h) and low temperatures (4 °C) result in acclimation, but low
temperatures produce a stronger response. An additive response is seen when low
temperatures and short days are experienced at the same time (Li et al. 2002).

Freezing tolerance of some herbaceous plant species such as Hordeum vulgare L.
cv. Dicktoo is also influenced by photoperiod and plants attain greater freezing tolerance
when exposed to 4 °C under short photoperiods (8 h) compared to long photoperiods (20
h). Plants grown under short days gradually increase in freezing tolerance and reach LTso
(temperature at which 50 percent of frozen plant crowns survive) of -12 °C after 56 d at 4
°C, while plants grown under long days peak at LT50 of -8 °C after 7 and then slowly
become less freezing tolerant. In this species, the increased freezing tolerance is
associated with a short-day induced delay in the transition from vegetative to
reproductive stage (Fowler et al. 2001). The roles that photoperiod and temperature play

in the acclimation of petunia (Petunia hybrida Vilm.) are currently unknown.

22

Electrolyte leakage assays are useful for comparing in vitro freezing tolerance
between different plant types or plants exposed to different environmental conditions
(Jaglo-Ottosen et al. 1998; Thapa et al. 2008; Pino et al. 2007; Pino et a]. 2008;
Pennycooke et al. 2003). This method provides a means of estimating cellular. membrane
damage by determining the percentage of total electrolytes lost due to freeze induced
membrane damage at various sub-zero temperatures. Typically, the temperature at which
50% of total electrolytes escape (ELso) is used for comparisons. Lower EL50
temperatures indicate a greater tolerance to freezing.

Garden petunias (P. hybrida) are very popular bedding plants, ranking first in
sales among bedding plants in 2007, with a wholesale value of over $111 million
(U SDA-NASS 2008). Estimates for the number of species belonging to the genus
Petunia vary, but numbers generally range from 11 (Kulcheski et al. 2006; Lorenze-
Lemke 2006) to 16 species (Griesbach 2007). All Petunia spp. originate from South
America (Kulcheski 2006) with the southeastern Sierra region of Brazil being one of the
centers of diversity (Lorenz-Lemke 2006). While the exact origin of P. hybrida is
unclear, it is believed to be the result of hybridization between two species, P. axillaris
Lam. and P. integrifolia Hook. (Sink 1984). Petunia hybrida is capable of cold
acclimation (Yelenosky and Guy 1989; Pennycooke et al. 2003).

Within the Solanaceae family, the genus Solanum encompasses species that vary
significantly in their ability to cold acclimate. The chilling-sensitive S. lycopersicum L.
(tomato) (Hsieh et al. 2002) and frost-sensitive S. tuberosum L. (potato) are species that
do not cold acclimate, while S. commersonii Dun. (a wild potato relative) is frost-tolerant

and capable of acclimation (Chen and Li 1980). Although S. tuberosum does not

23

normally cold acclimate in response to cool temperatures, heterologous expression of
AtCBF] or AtCBF 3 under a constitutive cauliflower mosaic virus (CaMV) 35S promoter
enhances the EL50 value of non-acclimated plants by about 2°C, from -3 °C to -5 °C.
When AtCBF] or AtCBF 3 are expressed behind a cold-inducible rd29A (Car 78)
promoter, non-acclimated plants modestly increase in freezing tolerance by 0.5-] °C, and
following acclimation at 2 °C for two weeks these lines show the same 2 °C increase in
freezing tolerance seen in the constitutively expressing lines (Pino et a1. 2007). In the
cold-acclimating S. commersonii, ectopic expression of AtCBF I behind a constitutive
promoter also increases freezing tolerance of non-acclimated plants by 4 °C. Unlike in S.
tuberosum, transgenic S. commersonii further increases in freezing tolerance following
two weeks at 2 °C (Pino et al. 2008). Heterologous expression of either AtCBF 3 or
LeCBF] (a CBF homolog from tomato) in S. lycopersicum does not result in a change in
freezing tolerance (Zhang eta]. 2004), but AtCBF] does confer an increase in tolerance
to low non-freezing temperatures (Hsieh et al. 2002).

During the processes of domestication and breeding, the genetic diversity of a
plant species tends to decrease dramatically (Tanksley and McCouch 1997). Strong
selection imposed by plant breeders, along with crossing between closely related
individuals, leads to a narrowing of the genetic diversity in future generations. For
example, soybean (Glycine max L.) is a crop with a very narrow range of genetic
diversity, and pedigree analysis has shown that just 35 genotypes contributed to 95% of
the germplasm base in modern cultivars (Gizlice et al. 1994).

Utilizing wild germplasm for crop improvement has proven to be an effective

method for increasing genetic diversity of crops. For example, quantitative trait locus

24

(QTL)-alleles transferred from the wild tomato species, Solanum habrochaites Knapp &
Spooner and S. pimpinellifolium L., to the cultivated tomato, S. lycopersicum L. by
advanced backcross methods improved horticultural traits in 22 out of 25 cases
(Bernacchi et al. 1998).

In the genus Petunia, interspeciflc hybridization is relatively easy to accomplish
(Ando et al. 2001; Griesbach 2007; personal experience), making utilization of wild
germplasm possible. Griesbach et a]. (1999) found that the red color of P. exserta results
from genetic factors distinct from those in current red colored P. hybrida cultivars.
Interspecific hybridization using P. exserta is adding to the genetic diversity of P.
hybrida (Griesbach et al. 1999). The simplest method of introducing wild germplasm
into a cultivated plant is via interspecific hybridization between a wild species and the
cultivated species. Hybrids from a segregating population exhibiting desirable traits are
then backcrossed using the cultivated species as the recurrent parent.

Improving cold tolerance of petunia would benefit both consmners and
commercial growers. The selection of bedding plants suited for growth in early spring
and late fall, when cold night temperatures and frosts still limit the survival of most
bedding plants, is limited. The objectives of the current study are 1) to examine the roles
of low temperature and photoperiod in cold acclimation of petunia and 2) to evaluate cold
acclimation of P. axillaris, P. integrifolia, P. exserta, and P. hybrida ‘Mitchell’, in an
effort to identify genetic material that may be useful in breeding more stress tolerant

cultivars.

25

MATERIALS AND METHODS:

Environmental cues influencing acclimation of P. hybrida ‘Mitchell’
Plant growth and accmation treatments

Seeds of P. hybrida ‘Mitchell’ were surface planted onto 288-cell plug trays filled
with 70% peat moss, 21% perlite, 9% vermiculite (v/v) (Suremix, Michigan Grower
Products Inc., Galesburg MI, USA). Trays were placed under intermittent overhead mist
irrigation until the development of at least two true leaves. Plants were then transferred
to a greenhouse at 22 °C with a 16 h photoperiod (ambient plus supplemental high-
pressure sodium lighting from 0600 to 2200 HR) until three weeks post planting.
Seedlings were then transplanted to 50-cell trays containing the same soilless media and
moved to a controlled environment growth chamber at 22 °C under a 16 h photoperiod
(100-130 umol m'2 s'1 fluorescent plus incandescent lighting) for one week. Fifty plants

were then subjected to each of seven different treatments shown in table 2.].

Cold tolerance determination by electrolyte leakage assay

Following temperature/photoperiod treatments, leaf discs from the upper portion
of the plant were collected using a 0.6 cm cork borer. The leaves chosen for sampling
were the youngest leaves that could be punched to obtain complete 0.6 cm diameter discs
without cutting into the midrib. Discs were immediately transferred to plastic weigh
boats filled with deionized water and stirred gently. Approximately 120 punches were
taken from a population of 50 plants for each treatment. Three discs were then
transferred to each of 30 (16x100mm) borosilicate glass culture tubes for each treatment

and placed on ice. When all discs were transferred, tubes were placed in a -1 °C

26

controlled temperature antifreeze bath (master bath) for 60 min. Three tubes of each
treatment were left on ice as controls. After 60 min, a small amount of ice was added to
each tube to nucleate extracellular ice formation. Tubes were plugged with foam and
kept at -1 °C for an additional 60 min, after which three tubes of each treatment were
moved to a second antifreeze bath at -1 °C, kept there for 40 min, and removed to ice.
Meanwhile, the temperature of the master bath was lowered to -2 °C. After 20 min, three
tubes of each treatment were moved from the master bath to another antifreeze bath at -2
°C for 40 min and then removed to ice (a total of 60 min at the test temperature). This
process continued for all temperatures tested (-1 °C to -9 °C). Tubes were then placed in
racks on top of ice and placed at 25°C to thaw slowly overnight.

The following day, 6 ml of dH20 was added to each tube followed by gentle
shaking for 3 h at room temperature to allow released electrolytes to diffuse into the
water. The water was then transferred to a new culture tube and electrical conductivity
(L.) was measured using a CON 110 conductivity meter (Oakton Instruments, Vernon
Hills IL, USA). Plant discs remained in the original tube and were frozen to -80 °C
overnight to release all electrolytes. The next day, the water from the previously
measured tubes was poured back into the corresponding plant disc tube, followed by
shaking for 3 h at room temperature. Conductivity of the water was again measured to
obtain the reading for maximum electrolyte leakage (L2). Percent of total electrolyte

leakage at each test temperature was calculated by (L1 -:— L2)* 100. Data analysis was

carried out using Sigmaplot (SPSS Inc., USA) and SAS (SAS Institute Inc., USA)
software. A sigmoidal curve was fitted to the leakage data for each treatment according

to the equation: y = a] + (a2 + (1 + exp(a3 - a4 * T»), where y is the average percent

27

electrolyte leakage of the three tubes at each temperature T, using the curve fitting
firnction of Sigmaplot. The initial parameters were specified as a1=0.1, a2=99.9, a3=0.]
and a4= 0.1 with constraints imposed such that a1>0 and 0<a2<100. The temperature at
which 50% of total electrolytes were released (ELso) was calculated by solving the

equation EL50 = ((a3 — log((a2 + (50 — a1))—1))+ a4) where the parameters a], a2, a3 and

a4 are given by the equation of the fitted curve. At least two EL50 values were calculated
for each treatment from two separate replications of the experiment. ANOVA analysis
and mean separations using Fisher’s LSD (a=0.05) were conducted on the EL50 values

using PROC GLM to compare treatment effects.

Assay of wild Petunia detached leaf freezing tolerance
Plant growth and acclimLfion of wild Petunia SDD.

Sowing and germination of P. integrifolia, P. exserta, P. axillaris (USDA
Accession 28546), P. axillaris (USDA Accession 28548) and P. hybrida ‘Mitchell’ were
as described above. When seedlings had two true leaves, plants were transferred to a
greenhouse at 20 °C with a 16 h photoperiod for four weeks. Seedlings were then
transplanted into 50-cell trays and moved to a controlled environment grth chamber at
22 °C, 16 h photoperiod (100-130 umol m"2 s'I light). After two weeks, plants of each
species were divided into two groups; a non-acclimated group remaining at 22 °C for an
additional week before testing, and an acclimated group subjected to a gradual cooling
process. The acclimated group was moved to 15 °C, 9 h photoperiod (100-130 umol m'2
3'1 light) for 1 week followed by 10 °C, 9 h photoperiod (100-130 umol rn'2 5'1 light) for 1

week, and finally 3 °C, 9 h photoperiod (30 umol m’2 5'1 light) for an additional week. At

28

the time of testing, non-acclimated plants were 7 weeks old and acclimated plants were 9
weeks old. However, plants in both groups had developed approximately the same
number of nodes at the time of sampling. Electrolyte leakage assays were performed as

described above.

RESULTS

Influences of environmental cues on cold acclimgtion of P. hybrida ‘Mitchell’

Temperature regime, but not photoperiod, influenced cold acclimation of P.
hybrida ‘Mitchell’ (Fig. 2.1 and 2.2; Table 2.2 and 2.3). When grown at 22 °C and 16 h
long days (NONLD), P. hybrida ‘Mitchell’ EL50 value was -1.9 °C (Fig. 2.1A and 2.2).
Exposing plants to three weeks of short days alone (N ONSD) did not increase freezing
tolerance and the EL50 temperature remained at -2.0 °C (Fig. 2. 1 B and 2.2). Both
NONLD and NONSD reached maximum electrolyte leakage at -4 °C (Fig. 2.1A and B).
Gradual cooling, whereby plants were grown under 16 h photoperiod for 7 d at each
temperature of 15 °C, 10 °C and 3 °C (RampdownLD) resulted in significant cold
acclimation, as the ELso temperature decreased to -5.1 °C (Fig. 2.1G and 2.2). The same
gradual cooling experienced under 9 h days (RampdownSD) also caused significant
acclimation but did not increase the acclimation response compared with RampdownLD
and the ELso temperature remained at -5.0 °C (Fig. 2.1F and 2.2).

Moving plants directly from long days at 22 °C to short days at 3 °C for just one
week (CA1) increased freezing tolerance slightly (ELso = -3.9°C) compared to NONLD
and NONSD (Fig. 2.1C and 2.2). Prolonging time of exposure to 3 °C with short days to

3 weeks (CA3) caused a more pronounced increase in freezing tolerance (ELSO = -4.9 °C)

29

compared to the CA1 treatment (Fig. 2.1E and 2.2). Maximum leakage for CA] occurred
at -7 °C compared with -8 °C for CA3 (Fig. 2.1C and E). Exposing plants to short days at
22 °C for two weeks prior to exposure to 3 °C with short days for another week (CA2),
gave an EL50 value of -4.0 °C and maximum leakage at -7 °C (Fig. 2.1D and 2.2), which
was similar to CA] (Fig. 2.2).

Exposure to three weeks of gradual cooling in both the RampdownSD and
RampdownLD treatments resulted in the same degree of acclimation as the CA3
treatment with three weeks at 3 °C (Fig. 2.2). However, the RampdownSD and
RampdownLD treatments were more effective than the CA1 treatment with just one week
at 3 °C (Fig. 2.2).

Freezing tolerance of wild Species

All species tested had similar constitutive freezing tolerance (Fig. 2.3 and 2.4;
Table 2.4A) with an ELSO temperature of ca. -2 °C without cold acclimation. Acclimated
ELso temperatures varied across species (Fig. 2.4; Table 2.4B and 2.5). The most cold-
tolerant species was P. axillaris (USDA accession 28548) with an acclimated ELSO
temperature of -7.8 °C, while P. exserta was the least cold-tolerant with an EL50
temperature of -4.9 °C. In addition to the observed interspecific variation, there was also
significant variation in the acclimated freezing tolerance between the two P. axillaris
accessions. The acclimated ELSO temperatures for P. hybrida, P. axillaris (USDA
accession 28546) and P. integrifolia are -6.0 °C, -6.1 °C and -6.6 °C, respectively. While
there was no statistical difference in the acclimated cold tolerance between these three

species, they were all significantly less cold tolerant that P. axillaris (accession 28548)

30

and all except P. axillaris (accession 28546) were significantly more cold tolerant than P.

exserta.

DISCUSSION AND CONCLUSIONS

Petunia hybrida is capable of cold acclimation (Yelenosky and Guy 1989;
Pennycooke et al. 2003 ), however, the environmental changes that signal the plant to
undergo acclimation have not previously been determined. The impacts of photoperiod
and cool temperatures were examined in this study. Previous work indicated that reduced
photoperiod and low temperatures both play a role in cold acclimation of woody trees
such as Pyrus malus (Howell and Weiser 1970) and Betula pendula (Li et al. 2002), and
in the acclimation of some non-woody plants such as Hordeum vulgare cv. Dicktoo
(Fowler et al. 200] ). Our results indicate that cool temperatures, but not reduced
photoperiods, influence cold acclimation of P. hybrida ‘Mitchell’ (Fig. 2.2). Plants
grown under long day conditions (N ONLD) showed similar basal freezing tolerance
levels as those exposed to a reduced photoperiod without a reduction in temperature
(NONSD), indicating that in the absence of cooling, short days alone are insufficient to
induce cold acclimation. Petunia hybrida ‘Mitchell’ did cold acclimate when exposed to
3 °C for one week along with a reduction in photoperiod from 16 h of light to 9 h (CA1),
increasing in freezing tolerance by ca. 2 °C. However, freezing tolerance did not further
increase with the addition of two weeks of exposure to short days prior to cold treatment
(CA2) (Fig. 2.2). Also, subjecting plants to gradual cooling had the same affect on cold
acclimation whether under long days (Rampdown LD) or short days (Rampdown SD).

The ELso value was ca. -4.9 °C following either treatment.

31

Duration at 3 °C influenced cold acclimation. One week at 3 °C (CA1) resulted in
an acclimated ELso temperature of -3.9 °C while 3 weeks (CA3) resulted in an acclimated
freezing tolerance of -4.9 °C. Similar results were reported for S. commersonii where
longer exposures to cool temperatures (2 °C) result in increased acclimation. In S.
commersonii, freezing tolerance gradually increases with duration of exposure until a
maximum is reached after 7 days (Pino et al. 2008). Likewise, in H. vulgare cv Dicktoo
maximum freezing tolerance is not reached until 56 days of acclimation at 4 °C under
short days (Fowler et al. 2001). The mechanisms involved in cold acclimation of petunia
are not fully understood and it would be of interest to determine the physiological and
molecular adaptations occurring during prolonged cooling.

One week at 3 °C preceded by a week at 15 °C and a week at 10 °C
(RampdownSD and RampdownLD) induced similar freezing tolerance as 3 weeks at 3 °C
(CA3). In A. thaliana, the level of CBF transcription is dependent upon the magnitude of
temperature change, with shifts to lower temperatures causing higher accumulation, but
sudden changes as slight as a drop from 20 °C to 10 °C are sufficient to induce CBF
transcription. With gradual cooling at a rate of 2 °C h", CBF transcripts become
detectable in A. thaliana at temperatures as high as 14 °C and continued cooling causes
even higher levels of transcript accumulation (Zarka et al. 2003). The initial weeks at 15
°C and 10 °C during the RampdownSD and RampdownLD regimes appear to be
sufficiently cool to induce petunia cold-acclimation pathways, allowing plants to
eventually reach the same level of freezing tolerance during the final week at 3 °C as was

reached following three weeks at 3 °C (CA3).

32

We examined the freezing tolerance of several wild Petunia species to identify
genetic sources of cold tolerance for use in future breeding. All of the tested species
showed similar freezing tolerance in the absence of cold acclimation, with an EL50 value
near -2 °C (Fig. 2.4). These results are consistent those of Pennycooke et al. (2003) who
reported an ELso value of ca. -2.5 °C on nonacclimated P. hybrida ‘Mitchell’ using a
slightly different electrolyte leakage assay procedure. The nonacclimated freezing
tolerance of P. hybrida ‘Mitchell’ is similar to S. lycopersicum with an ELso temperature
of -2 °C (Zhang et al. 2004), and S. tuberosum with an EL50 of -3 °C (Pino et al. 2007).
The frost-resistant species S. commersonii and A. thaliana ecotype RLD have
nonacclimated ELSO temperatures of -5 °C and -4 °C, respectively (Pino et al. 2008; Jaglo-
Ottosen et al. 1998). These comparisons suggest that Petunia is relatively sensitive to
freezing prior to acclimation.

Although all Petunia species showed similar nonacclimated freezing tolerance,
significant variation was seen following acclimation. The extent of acclimation ranged
from only a 3.0 °C increase in freezing tolerance for P. exserta to 5.8 °C for P. axillaris
(accession 28548). It is interesting to note that although P. axillaris (accession 28548)
was the most cold-tolerant plant that we tested, our other accession of the same species,
P. axillaris (accession 28546), was not statistically different than P. exserta and increased
in freezing tolerance by 4.3 °C. These two accessions of P. axillaris showed significant
morphological differences as well, with accession 28548 having a much longer corolla
tube and leaves than accession 28546 (data not shown). Although the exact collection
site of these two accessions is unknown, they are likely different ecotypes. Accession

28546 may have come from a warmer environment than accession 28548, and therefore

33

environmental pressures have resulted in the selection of more cold tolerant individuals in
accession 28548.

Three of the species, P. integrifolia, P. hybrida and P. axillaris (acccession
28546), had freezing tolerances that are not statistically different following acclimation,
with an ELso near -6 °C. The modern garden petunia, P. hybrida, is believed to have
arisen from the hybridization of P. integrifolia and P. axillaris. We also know that the P.
hybrida ‘Mitchell’ cultivar used in these tests arose from the crossing of P. axillaris X (P.
axillaris X P. hybrida ‘Rose du Ciel’) (Griesbach 2007). Therefore, it is not surprising
that P. hybrida ‘Mitchell’ has cold tolerance characteristics that are so similar to these
two putative parental species.

As a point of comparison between petunia and other species, neither S.
lycopersicum (Zhang et al. 2004) nor S. tuberosum (Pino et a1. 2007) increase in freezing
tolerance following cold acclimation. On the other hand, freezing-tolerant S.
commersonii (Pino eta]. 2008) and A. thaliana ecotype RLD (Jaglo-Ottosen et al. 1998)
both increase in freezing tolerance by ca. 4 °C to reach acclimated ELso temperatures of -
9 °C and -8 °C, respectively. Therefore, we conclude that Petunia, and especially P.
axillaris (accession 28548) with an acclimated EL50 temperature near -8 °C (an increase
in tolerance of 5.8 °C over nonacclimated plants), has the ability to cold acclimate by a
significant amount for an herbaceous species.

The results of these experiments show that there is significant variation in the
freezing tolerance of Petunia spp. following a period of cold acclimation. This variation
could be utilized when breeding new Petunia cultivars by incorporating germplasm from

the more cold tolerant wild species, P. axillaris (accession 28548), into the breeding

34.

program. Further work is necessary to identify genetic differences controlling this
variation in acclimation ability. Cold tolerance is a quantitative trait expected to be
controlled by several genetic factors. Identification of QTL underlying cold acclimation
ability in Petunia would be useful for implementing marker assisted breeding programs
aimed at efficiently transferring cold tolerance traits from the wild species to the

cultivated plants.

35

Table 2.1. Growth conditions (temperature, photoperiod, photosynthetic photon flux
(PPF) and duration) of P. hybrida 'Mitchell' glantsprior to electrolyte leakage assays.

 

 

Temperature Photoperiod PPF Duration

Growth regimes (°C) (b) (umol m'2 s") (weeks)
NONLD 22 16 100-130 6
NONSD 22 16 100-130 4

22 9 100-130 3
CA] 22 16 100-130 6

3 9 30 1
CA2 22 16 100- 1 30 4

22 9 100-130 2

3 9 3O 1
CA3 22 16 100-130 5

3 9 30 3
RampdownSD 22 16 100-130 5

15 9 100-130 ]

10 9 100-130 1

3 9 30 1
RampdownLD 22 16 100-130 5

15 16 100-130 1

10 16 100-130 1

3 16 30 1

 

Table 2.2. ANOVA for effect of acclimation regime on freezing tolerance of P.
hybrida 'Mitchell' as measured by electrolyte leakage assays.

 

 

Dependent variable: Sum of Mean F

ELso temperature Source DF Squares Square Value Pr > F
Model 6 22.4 3 .73 21.3 0.0004
Error 7 1.23 0.18
Total 13 23.6

 

Table 2.3. Pairwise comparisons with Fisher's LSD between ELSO values of P.
hybrida ‘Mitchell’ grown under different acclimation regimes. Shown are p-
values for each comparison.

 

NONLD NONSD CA1 CA2 CA3 RampdownSD
NONSD 0.810
CA] 0.002 0.003
CA2 0.002 0.002 0.721
CA3 0.000 0.000 0.036 0.062
RampdownSD 0.000 0.000 0.035 0.060 0.978
RampdownLD 0.000 0.000 0.022 0.038 0.744 0.765

 

36

Table 2.4. ANOVA for effect of species on nonacclimated (A) and acclimated (B)
ELso temperature as determined by electrolyte leakage assay.

 

 

 

 

A. Dependent variable: Sum of Mean F

EL50 temperature Source DF Squares Square Value Pr > F
Model 4 0.12 0.03 0.07 0.99
Error 10 4.47 0.45
Total 14 4.59

B. Dependent variable: Sum of Mean F

EL50 temperature Source DF Squares Square Value Pr > F
Model 4 15.6 3.89 9.63 0.002
Error 10 4.04 0.40
Total 14 19.6

 

Table 2.5. Pairwise comparisons with Fisher's LSD between acclimated
EL50 values of different Petunia species. Shown are p-values for each
comparison.

 

P. axillaris P. axillaris

(28546) (285 48) P. exserta P. mtegrzfolza

P. axillaris

(28548) 0009
P. exserta 0.080 0.000
P. integrifolia 0.439 0.044 0.018
P. hybrida
'Mitchell' 0.967 0.003 0.046 0.353

 

37

 

 

 

 

 

 

 

 

 

5’? F‘zffi 5‘: via “*1 ,1. g’
t; l 2!“: at .r . l x
'3‘: {_‘j' {.1 E j i' .l ., I‘ 80 m
1;? l3; £73. l'jl t ' 0
:6 life life iii-,2. til-24 if: j 60 ,\°
t 8.
.. :12? l E " s

L5.- >'>-,: . 40 g
L5“ 2""; t: -:.’. ," ’

j l l 20 g
‘13-'- l'I-'- .-' '4

‘.-,~ w I; ‘ ‘33! S, r 0

  

 

 

CAT c]

i
1" i; ii

4 14 ' >
. "r. MP" W..-
. ‘ _ . . o_ ' . . .

Average % Leakage

 

I‘ . I .l.’. I '_
L.I_\J.._._~J..-.n_" -L- - ._
4 .- ._._r,,,... .. -—-
..
’...- ",
... '. "
- . ,. .'_ A

 

 

' in. J.,. -
_, . _ t... '. V I . . . .-

A'M‘W - “s...“ ' .:‘:-;

(m.

..

 

 

7
. \

§
e

.1-4—’-;‘A.---AL~‘J t...

'le
on
o

V.t'- r... .__
. g _.
, .. .

5 L 3::
c'»
L
.5
0
Average % Leakag

- —'--.- WVm—Pk‘ .-. a 7.- ' I- .
- ' '. ' ._ .
.. - . r .
., ~ . v. - _. .
. . I - . - .
. _ - 1' . a .

 

I
“#4.: MA

 

 

 

 

   

Temperature (°C)

 

RampdownLD G

 

 

 

Temperature (°C)

 

‘ 120
. 100 0

'1‘ .

‘I ,
\1

-‘ 1‘

.l ‘1
...,.;

 

--_ w—fi—fi
r‘ .5.-v.‘
. - ‘ '
‘.. v
I a.
‘-

Average % Leakag

”.57“?
.'_
77-1‘_

    

 

 

 

 

 

 

 

 

 

 

 

 

-1O -8 -6 -4 -2
Temperature (°C)

 

CA2 D 120

[—4

.5
0

LI.— '
0 ’L;_.'i‘
o L:
O

 

WW?
0)
0

Average % Leakag

 

. . . - we I
. . l .‘g' . .
AA-......M_._ -u...o.,-.»..‘._‘_..J

‘k‘._—v-A-. -
_.._._.__—..

 

_. .w .—.—.—--.......~
I

i. —i
.‘I .' '_ ._'I__-.1

 

 

 

-6 -4 -2
Temperature (°C)

 

RampdownSD F

 

 

 

-10 -8 -6 -4 -2 0
Temperature (°C)

Figure 2.1. Percentage of electrolyte leakage measured
at various freezing temperatures on leaf discs of P.
hybrida ‘Mitchell’ grown under different conditions. (A)
nonacclimation under long days (NONLD), (B)
nonacclimation under short days (NONSD), (C) cold
acclimation regime ] (CA1), (D) cold acclimation regime
2 (CA2), (E) cold acclimation regime 3 (CA3), (F)
rampdown under short days (Rampdown SD), (G)
rampdown under long days (Rampdown LD). NONLD:
22°C, LD (16hr days). NONSD: 22°C, LD; then 3 weeks
at 22°C, SD (9hr days). CA]: 22°C, LD; then 1 week at
3°C, SD. CA2: 22°C, LD; then 2 weeks at 22°C, SD;
then 1 week at 3°C, SD. CA3: 22°C, LD; then 3 weeks at
3°C, SD. RampdownSD: 22°C, LD; then 1 week at 15°C,
SD; then 1 week at 10°C, SD; then ] week at 3°C, SD.
Rampdown LD: 22°C, LD; then 1 week at 15°C, LD;
then 1 week at 10°C, LD; then 1 week at 3°C, LD. Error
bars represent standard deviation of six measurements.

38

 

 

(d) WWW RampdownLD
(ed) WWW RampdownSD
(ed) WWW CA3

(be) WWW CA?-

G» %’////////////////////////////////// CA1
(a) %W NONSD
(3) %%////////fl NONLD

 

 

-7 -6 -5 -4 -3 -2 -1 0
EL50 Temperature (°C)

Figure 2.2. EL50 values for P. hybrida ‘Mitchell’ after various cold acclimation regimes.
NONLD: 22°C, LD (16hr long days). NONSD: 22°C, LD; then 3 weeks at 22°C, SD (9hr
short days). CA1: 22°C, LD; then 1 week at 3°C, SD. CA2: 22°C, LD; then 2 weeks at
22°C, SD; then 1 week at 3°C, SD. CA3: 22°C, LD; then 3 weeks at 3°C, SD.
RampdownSD: 22°C, LD; then 1 week at 15°C, SD; then 1 week at 10°C, SD; then 1
week at 3°C, SD. RampdownLD: 22°C, LD; then 1 week at 15°C, LD; then 1 week at
10°C, LD; then 1 week at 3°C, LD. Bars with the same letter are not statistically different
according Fisher’s LSD with a = 0.05. Error bars represent standard deviation of the two
EL50 values calculated from two replications for each treatment.

39

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

120

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A Petunia hybrida 'Mitchell'
.. . t 1003,
i n=l2 r 80 %
CD
A
l 60 S”
8,
"40 g
9
l i ll 4° “
- [I o
-10 -8 -6 -4 -2 0
Temperature (°C)
. . . 120
C Petunia aXIIIans (USDA 28548)
i 1003,
_ E
- n—I2L 80 8
_l
t 60 °\°
8)
t 40 I3
. 9
i l I I ll "° “
- , . 0
-10 -8 -6 -4 -2 0
Temperature (°C)
.5 Petunia integrifolia 120
.. l 100%
ii 11' _ x
n—6 . 80 8
_I
l 60 ..\°
8
>40 3
9
E L 20 <
. . o
-10 -8 -6 -4 -2 0

Temperature (°C)

 

B Petunia axillaris (USDA 28546)

“F

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

":6 t-

’20

120
~ 100.»
so
loo
»40

Average % Leakag

 

 

-8 -6 4 -2

Temperature (°C)

 

Petunia exserta

 

 

-r

m-.-.-. -; ,- m
._-..
- l . l I - ‘
um... E...'.—.._ ha I...: “4.: -;~.:.£.Luw.'J—TA— W
I
‘
_ .— .. . .

 

 

 

 

 

 

 

 

 

 

 

m.-..:-::;.L.L"n-a:n—-im -

 

n:

120
t 100
. 80
t 60
t 40
t 20

Average % Leakage

 

 

-6 -4 -2
Temperature (°C)

0

 

- Non acclim ated
[:3 Acclimated

 

 

 

Figure 2.3. Average percent electrolyte leakage at each temperature tested for Petunia
hybrida (A) and wild Petunia species (B-E). Nonacclimated plants were grown at 22°C
LD and acclimation was accomplished by exposing plants to 15°C SD for 1 week, 10°C

SD for 1 week, then 3°C SD for 1 week. Leakage data at each temperature point is

averaged over n measurements. Standard deviation shown by error bars.

40

 

 

(c) Saw ,3- axillaris 23543

(b) r: P, ,,,,,g,,-,,,,,

0)) Raw p. hybrida 'Mitchell'

r , ‘ ._.
.- -

r

(ab) ::::m\\\\\\\\\\\\\\\\\\\\\x P. ,x,,,,,,-, 2,54,

-m ’
s..- .

(a) n:\\\\\\\\\\\\\\\ \\\\V p, we...

    

“

 

 

1.; 3.

- Nonacclimated
Acclimated
-10 -8 -6 -4 '2 0
ELso Temperature (Celsius)

 

 

 

-I

Figure 2.4. ELSO temperatures for nonacclimated (7 weeks at 22°C LD) and acclimated
(6 weeks at 22°C LD, 1 week at 15°C SD, 1 week at 10°C SD, and 1 week at 3°C SD)
wild Petunia species. Bars depict EL50 temperatures averaged over at least 2 replications
for each species. Bars with the same letter are not statistically different according
Fisher’s LSD with a = 0.05. Standard deviation shown by error bars.

41

REFERENCES

Ando T., M. Nomura, J. Tsukahara, H. Watanabe, H. Kokubun, T. Tsukamoto,
G. Hashimoto, E. Marchesi, and I. Kitching. 2001. Reproductive isolation in a
native pOpulation of Petunia sensu Jussieu (Solanaceae). Ann. Bot. 88:403-413.

Bemacchi D., T. Beck-Bunn, D. Emmatty, Y. Eshed, S. Inai, J. Lopez, V. Petiard,
H. Sayama, J. Uhlig, D. Zamir, and S. Tanksley. 1998. Advanced backcross
QTL analysis of tomato. 11. Evaluation of near-isogenic lines carrying
single-donor introgressions for desirable wild QTL-alleles derived from
Lycopersicon hirsutum and L. pimpinellifolium. Theor. Appl. Genet.

97: 170-180.

Chen H.H., and RH. Li. 1980. Characteristics of cold acclimation and deacclimation in
tuber-bearing Solanum species. Plant Physiol. 65:]146-1148.

Cook D., S. Fowler, 0. Fiehn, and M. Thomashow. 2004. A prominent role for the CBF
cold response pathway in configuring the low-temperature metabolome of
Arabidopsis. Proc. Natl. Acad. Sci. 101:15243-15248.

Fowler D.B., G. Breton, A.E. Limin, S. Mahfoozi, and F. Sarhan. 2001. Photoperiod and
temperature interactions regulate low-temperature-induced gene expression in
barley. Plant Physiol. 127: 1676-1681.

Gilmour S., D. Zarka, E. Stockinger, M. Salazar, J. Houghton, and M. Thomashow. 1998.
Low temperature regulation of the Arabidopsis CBF family of AP2 transcriptional
activators as an early step in cold-induced COR gene expression. Plant J. 16:433-

442.

Gilmour S., S. Fowler, and M. Thomashow. 2004. Arabidopsis transcriptional activators
CBF], CBF 2, and CBF 3 have matching functional activities. Plant Mol. Biol.
54:767-781.

Gizlice Z., T.E. Carter, Jr., and J .W. Burton. 1994. Genetic base for North American
public soybean cultivars released between 1947 and 1988. Crop Sci. 34:1143-
1151.

Griesbach R.J., J .R. Stehmann, and F. Meyer. 1999. Anthocyanins in the “red” flowers of
Petunia exserta. Phytochemistry 51:525-528.

Griesbach R.J. 2007. Petunia, p. 301-336. In: N.O. Anderson (ed.). Flower breeding and
genetics: Issues, challenges and opportunities for the 21 st century. Springer,
Netherlands.

Guy C., K. Niemi, and R. Brambl. 1985. Altered gene expression during cold acclimation
of spinach. Proc. Natl. Acad. Sci. 82:3673-3677.

42

Hsieh T., J. Lee, P. Yang, L. Chiu, Y. Charng, Y. Wang, and M. Chan. 2002a.
Heterology expression of the Arabidopsis C—Repeat/Dehydration Response
Element Binding Factor 1 gene confers elevated tolerance to chilling and
oxidative stresses in transgenic tomato. Plant Physiol. 129:1086-1094.

Howell GS. and OJ. Weiser. 1970. The environmental control of cold acclimation
in apple. Plant Physiol. 45:390-394.

Jaglo-Ottosen K., S. Gilmour, D. Zarka, O. Schabenberger, and M. Thomashow. 1998.
Arabidopsis CBF] overexpression induces COR genes and enhances freezing
tolerance. Science 280: 104-106.

Kulcheski F.R., V.C. Muschner, A.P Lorenz-Lemke, J .R. Stehmann, S.L. Bonatto,
F.M. Salzano, and LB. F reitas. 2006. Molecular phylogenetic analysis of
Petunia Juss. (Solanaceae). Genetics 126:3-14.

Li C., T. Puhakainen, A. Welling, A. Vihera—Aarnio, A. Emstsen, O. Juntilla, P. Heino,
and ET. Palva. 2002. Cold acclimation in silver birch (Betula pendula).
Develpoment of freezing tolerance in different tissues and climatic ecotypes.
Physiol. Plant. 116:478-488.

Lorenz-Lemke A.P., G. Mader, V.C. Muschner, J .R. Stehmann, S.L. Bonatto, F.M.
Salzano, and LB. Freitas. 2006. Diversity and natural hybridization in a highly

endemic species of Petunia (Solanaceae): A molecular and ecological
analysis. Mol. Ecol. 15:4487-4497.

Pennycooke J ., M. Jones, and C. Stushnoff. 2003. Down-regulating u-Galactosidase
enhances freezing tolerance in transgenic petunia. Plant Physiol. 133:901-909.

Pino M.T., J .S. Skinner, E.J Park, Z. Jeknic, P.M. Hayes, M.F. Thomashow and T.H.H.
Chen. 2007. Use of a stress inducible promoter to drive ectopic AtCBF expression
improves potato freezing tolerance while minimizing negative effects on tuber
yield. Plant Biotechnol. J. 5:591-604.

Pino M.T., J .S. Skinner, Z. Jeknic, P.M. Hayes, A.H. Soeldner, M.F. Thomashow, and
T.H.H. Chen. 2008. Ectopic AtCBF] over-expression enhances freezing tolerance
and induces cold acclimation-associated physiological modifications in potato.
Plant Cell Environ. 31:393-406.

Sink KC. 1984. Taxonomy, p. 3-9. In: Sink KC (ed.). Monographs on theoretical and
applied genetics 9: Petunia. Springer-Verlag, Berlin.

Steponkus P.L., M. Uemura, R.A. Joseph, S. Gilmour, and M. Thomashow. 1998. Mode
of action of the COR15a gene on the freezing tolerance of Arabidopsis thaliana.
Proc. Natl. Acad. Sci. 95:14570-14575.

43

Stockinger E., S. Gilmour, and M. Thomashow. 1997. Arabidopsis thaliana CBF]
encodes an AP2 domain-containing transcriptional activator that binds to the C-
repeat/DRE, a cis-acting DNA regulatory element that stimulates transcription in
response to low temperature and water deficit. Proc. Natl. Acad. Sci. 94:1035-
1040.

Tanksley SD. and SR. McCouch. 1997. Seed banks and molecular maps: Unlocking
genetic potential from the wild. Science 277:1063-1066.

Thapa B., R. Arora, A. Knapp, and E. Brummer. 2008. Applying freezing test to quantify
cold acclimation in Medicago truncatula. J. Amer. Soc. Hort. Sci. 133:684-691.

Thomashow M. 1999. Plant cold acclimation: Freezing tolerance genes and regulatory
mechanisms. Annu. Rev. Plant Physiol. Plant Mol. Biol. 50:571-599.

United States Department of Agriculture (USDA)-National Agricultural Statistics
Service (NASS). 2008. Floriculture Crops 2007 Summary.
<http://usda.mannlib.comell.edu/usda/current/FlorCrop/FlorCrop-04-24-
2008.pdf>

Yelenosky G. and C. Guy. 1989. Freezing tolerance of citrus, spinach, and petunia leaf
tissue. Plant Physiol. 89:444-451.

Zhang X., S. Fowler, H. Cheng, Y. Lou, S. Rhee, E. Stockinger, and M. Thomashow.
2004. Freezing-sensitive tomato has a functional CBF cold response pathway, but
a CBF regulon that differs from that of freezing-tolerant Arabidopsis. Plant J.
39:905-919.

44

CHAPTER 3

EVIDENCE FOR CONSERVATION OF THE CBF COLD-RESPONSE
PATHWAY AND ITS ROLE IN COLD ACCLIMATION OF PE T UNIA H YBRIDA
‘MITCHELL’

ABSTRACT

Freezing tolerance of many plant species increases following exposure to low,
non-freezing temperatures, a process termed cold acclimation. In the model plant species
Arabidopsis thaliana, the CBF family of transcriptional activators plays an important role
in cold acclimation. The CBF-regulon is conserved in a number of species including
members of the family Solanaceae. The objectives of this work were to determine
whether heterologous expression of CBF genes improves freezing tolerance of Petunia
hybrida and whether a functional CBF regulon is present in petunia. Petunia hybrida
‘Mitchell’ plants were transformed with either AtCBF 3 from Arabidopsis or LeCBF]
from Solanum lycopersicum, both under the strong constitutive CaMV 35S promoter. A
single transgenic line with high expression levels of AtCBF 3 showed an increase in
constitutive freezing tolerance, while acclimated tolerance was similar to wild-type.
None of the LeCBF] -expressing lines showed any significant improvements in
constitutive or acclimated freezing tolerance. Expression of putative petunia CBF
transcription factors (petCBF1-4) was induced in response to chilling at 3 °C, and two
putative downstream genes of the petunia CBF pathway exhibiting increased expression

in response to both cold treatment and CBF over-expression were identified.

45

INTRODUCTION

Freezing temperatures can cause tremendous damage to plant cells, primarily to
the cellular membranes (Thomashow 1999). Plants vary greatly in their ability to tolerate
cold temperatures and are categorized as chilling sensitive, freezing sensitive, or freezing
tolerant. Chilling sensitive species suffer damage when exposed to cool, nonfreezing
temperatures. Freezing sensitive species suffer damage when temperatures drop below
freezing. Freezing tolerant species are able to tolerate temperatures below 0 °C.

Cold tolerance is a dynamic characteristic controlled by genetic and
environmental factors. When exposed to low non-freezing temperatures, many plant
species undergo a process termed cold acclimation which allows them to transiently
survive previously lethal low temperatures. During the process of cold acclimation,
changes occur in gene expression (Guy et al. 1985) and the cellular metabolite profile
(Cook et al. 2004). Some of the genes that are upregulated during cold acclimation serve
to stabilize cellular membranes (Steponkus et al. 1998). Other cold-regulated genes
encode antifreeze proteins, signal transduction proteins, and transcription factors
(Maruyama et al. 2004, Vogel et al. 2005), or act to increase concentrations of
compatible solutes such as proline and sugars in the cytoplasm (Gilmour et al. 2000).

In Arabidopsis thaliana (L.) Heynh., the cold-induced transcriptional activator
CBF] (Q-repeat/Dehydration responsive element binding factor 1) is largely responsible
for changes that occur during cold acclimation (Fowler and Thomashow 2002). CBF]
belongs to a small family of transcriptional activators in Arabidopsis consisting of two
other apparently functionally redundant genes, CBF 2 and CBF 3 (Gilmour et a]. 1998,

2004) located in tandem array on chromosome IV. The CBF proteins bind to CRT/DRE

46

(Q-repear/Dehydration responsive glement) sequences within the promoter regions of
cold regulated genes (Stockinger et al. 1997; Gilmour et al. 1998). CBF 1 , CBF 2, and
CBF3 proteins all have a mass of 24kDa, and are 88% identical at the amino acid level.
CBF transcripts begin to accumulate within 15 minutes of exposure to 2.5 °C. Transcript
levels continue to increase for two hours and then begin to slowly decline (Gilmour et al.
1998)

A robust set of low temperature-responsive genes, termed the COS (_CQld
Standard) set, that are reliably up- or down-regulated in response to low temperature,
have been identified in Arabidopsis (Vogel et al. 2005). Of the 513 genes comprising the
COS set, 93 are under CBF control (Fowler and Thomashow 2002). Among the COS
genes are members of the COR (COM-Responsive) family of genes that are induced in
Arabidopsis in response to cold temperatures (Baker et al. 1994; Thomashow 1999).
CORISa is involved in the stabilization of membranes during cold temperatures
(Steponkus et al. 1998; Thomashow 1999) and transcripts for COR15a along with
COR 78 begin to accumulate after two hours of cold exposure (Gilmour et al. 1998).

Constitutive expression of CBF] in Arabidopsis also induces COR gene
expression without a low temperature exposure. This increase in COR expression results
in an increase in the freezing tolerance of nonacclimated plants (Jaglo-Ottosen et al.
1998). Overexpression of CBF 3 in transgenic Arabidopsis results in plants that are more
salinity, drought, and cold tolerant compared to wild-type plants (Kasuga et al. 1999).
However, there are often undesirable phenotypic effects from placing CBF genes under
the control of the constitutive CaMV 35S promoter, including stunted growth, a more

prostrate growth form, shorter petioles, a bluish tint, and delayed flowering (Kasuga et al.

47

1999; Gilmour et al. 2000). When the CBF 3 gene was expressed under the control of the
stress inducible rd29A (Cor78) promoter, plants exhibit some stunting but not to the
extent present in the CaMV 35S plants. Transgenic plants in which the inducible
promoter is used are still more cold tolerant than control. plants following cold
acclimation (Kasuga et al. 1999).

Functional CBF genes have been identified in numerous species. A CBF ortholog
isolated from Zea mays L. (ZmDREBIA) induces expression of CBF target genes and
increases the freezing tolerance of nonacclimated Arabidopsis when expressed behind the
constitutive CaMV 35S promoter (Qin et al. 2004). Constitutive expression of the CBF
ortholog CIG-B from Prunus avium L. induces expression of COR] 5a when expressed in
Arabidopsis. These transgenic plants are more cold and salt tolerant than wild type plants
(Kitashiba et al. 2004). The freezing tolerance of several crop species, including
Brassica napus L. (Jaglo et al. 2001), Fragaria xananassa Duchesne (Owens et a]. 2002)
and Solanum tubersosum (Pino et al. 2007; Pino et al. 2008), has been enhanced by
heterologous expression of Arabidopsis CBF genes.

The Solanaceae family comprises members that span the spectrum of cold
tolerance (Chen and Li 1980). Tomato (S. lycopersicum L.) is a chilling sensitive
member of the family that is incapable of cold acclimation (Hsieh et al. 2002a). Potato
(S. tuberosum L.) is a freezing sensitive species lacking the ability to cold acclimate
(Chen and Li 1980; Pino et al. 2007). A wild potato relative, S. commersonii Dun., is a
frost resistant species that does cold acclimate (Chen and Li 1980). Garden petunias
(Petunia hybrida Vilm.) are a moderately cold tolerant species with the capacity to cold

acclimate (Yelenosky and Guy 1989; Pennycooke et al. 2003).

48

Tomato encodes three CBF homologs (LeCBF1-3) (Zhang et al. 2004). All three
LeCBF genes are induced by mechanical agitation but not by ABA, salinity, or drought.
LeCBF] is the only one of the three induced by low temperatures (Zhang eta]. 2004).
While CBF genes are present in tomato, plants do not cold acclimate. Freezing tolerance
does not increase when LeCBF] or AtCBFI is overexpressed in tomato (Zhang et al.
2004), but chilling and drought tolerance increases with AtCBF] overexpression (Hsieh
et al. 2002a,b). Overexpression of LeCBF] in Arabidopsis leads to COR gene induction
and increased freezing tolerance. However, in tomato, only four genes are substantially
induced by overexpression of either CBF gene (Zhang et al. 2004). Therefore, tomato
does encode a CBF regulon, but it is insufficient for cold acclimation.

Both freezing-sensitive S. tuberosum and freezing-tolerant S. commersonii are
responsive to heterologous expression of Arabidopsis CBF genes (Pino et al. 2007; Pino
et al. 2008). Constitutive expression of AtCBF 1 or AtCBF 3 in potato increases freezing
tolerance from -3 °C to -5 °C without a cold treatment. Additionally, expression of these
genes behind a cold-inducible rd29A promoter increases the freezing tolerance to -5°C
following a two-week treatment at 2 °C (Pino et al. 2007). When AtCBF 3 is
constitutively expressed in S. commersonii, freezing tolerance of warm grown plants
increases by 2 to 4 °C. Cold acclimation of these transgenic lines results in further
increases in freezing tolerance by anywhere from 1 to 4 °C (Pino et al. 2008).

Together, these studies show that the CBF pathway is highly conserved in the
family Solanaceae, yet cold tolerance varies widely. Although endogenous CBF
transcription factors are present in tomato, the downstream components required for cold

acclimation seem to be lacking. In potato, CBF overexpression enhances freezing

49

tolerance, indicating that downstream components of the regulon are present and
functional. In S. commersonii, CBF overexpression enhances freezing tolerance, but the
plants are able to acclimate further following cold treatment, suggeSting that other
pathways independent of CBF are also responsible for the cold tolerance of this species.
The conservation of the CBF regulon and the role it may play in the cold tolerance of P.
hybrida is not yet known. While the ability to cold acclimate demonstrates that the
cellular machinery for acclimation exists in petunia, the genetics controlling this ability
remain to be determined.

The objective of the present study was to determine whether Petunia hybrida has
a functioning CBF cold-response pathway. This was accomplished by determining
whether heterologous expression of CBF genes from Arabidopsis or S lycopersicum
confers an increase in freezing tolerance when expressed behind a constitutive CaMV 35S

or a cold-inducible AtC or] 5a promoter in petunia.

MATERIALS AND METHODS

Plant transformation and characterization
Transformgtion of Petunia hybrida ‘Mitchell’

Genetic transformation of Petunia hybrida ‘Mitchell’ was carried out by the Plant
Biotechnology Resource and Outreach Center at Michigan State University.
A grobacterium-mediated transformation on leaf explants was used to integrate one of
four CBF-containing constructs into petunia. Empty vector and non-transformed
explants were also carried through tissue culture as negative controls. See Figure 3.1 for

an explanation of constructs used in transformation. The pMPS]3 (Gilmour et al. 2000)

50

and pXIN 1 (Zhang et al. 2004) constructs were kindly provided by Dr. Mike Thomashow
(Michigan State University) and the pSPUD constructs were kindly provided by Dr. Dave
Douches (Michigan State University). Two separate rounds of transformation were
performed.
PCR of putative transgenic plants for transgene integration

DNA from all putative To generation transgenic plants was collected onto F TA
PlantSaver Cards (Whatman, United Kingdom) for use in PCR. A young expanding leaf
was detached from the plant and placed with the abaxial surface facing the card. A small
piece of plastic film was placed over the sample while it was pressed onto the card with a
pestle. The cards were air-dried overnight before use in a PCR reaction. For each
reaction, a 1 mm diameter punch was removed from the FTA card and placed into a thin-
walled PCR tube. The disc was washed twice for 5 min with 200 u] FTA Purification
Reagent (Whatman, United Kingdom) followed by two washes with 200 pl TE” (10 mM
tris-hydrochloride (Tris), 0.1 mM ethylenediaminetetraacetic acid (EDTA) pH 8.0).
Discs were dried for 20 min at 5 6 °C and cooled to room temperature prior to the addition
of a PCR master mix (1X Taq buffer, 0.2 mM dNTPs, 1 uM forward primer, 1 uM
reverse primer, 2.5 mM magnesium chloride (MgClz), 1 unit GoTaq Flexi DNA
polymerase (Promega, Madison WI, USA)) in 50 ul reactions. Twenty microliters of the
reaction product was separated on a 1% agarose gel for visualization. Table 3.] lists
primer sequences and PCR reaction conditions.
Southern hybridization analysis

Southern hybridization was performed to determine transgene copy number and to

confirm PCR results for presence of the transgenic construct. DNA was obtained using a

51

cetyl trimethylammonium bromide (CTAB) extraction method. Fresh (~2 g) or freeze
dried (~300 mg) plant tissue was ground in the presence of ~10 mg of polyvinyl
pyrrolidone (PVP mw 40,000) and transferred to a 15 ml conical tube with 5 ml CTAB
extraction buffer (2% CTAB, 100 mM Tris pH 8.0, 1.4 M (sodium chloride) NaCl, 20
mM EDTA pH 8.0, 1% B-mercaptoethanol (BME)). Samples were incubated at 65 °C for
l h with intermittent agitation. Samples were then extracted twice with an equal volume
chloroformziso-amyl alcohol (24:1) and precipitated with 1/10 volume 3 M sodium
acetate and either 2 volumes ethanol or 1 volume isopropanol at -20 °C. DNA was
pelleted, washed once with 70% ethanol, dried, resuspended in 500 pl TE and incubated
with 5 u] DNase-free RN ase (Roche Applied Science, Germany) for 1 h at 37 °C. DNA
was again precipitated with 1/10 volume 3 M sodium acetate and 2 volumes ethanol at -
20 °C. DNA was pelleted, washed twice with 70% ethanol, dried and resuspended in
sterile deionized water. Quantification was carried out using a BioSpec-mini UV-Visible
spectrophotometer (Shimadzu Corporation, Japan).

Total DNA of all transgenic lines was digested with either PstI or EcoRI (New
England Biolabs, Beverly MA, USA) restriction enzymes according to manufacturer’s
recommendations. A second blot of most transgenic lines was performed using a double
digest with EcoRI (New England Biolabs, Beverly MA, USA) and HindIII (Invitrogen
Corp, Carlsbad CA, USA) restriction enzymes to verify the results. Digested DNA (10
to 15 ug) was separated on a 1% agarose gel at 70 volts for 4 h. DNA was transferred to
a Hybond N+ (GE Healthcare, Buckinghamshire, United Kingdom) membrane by
overnight capillary transfer with 10X SSC (1.5 M NaCl, 0.15 M sodium citrate). DNA

was affixed to the membrane by UV crosslinking (Stratalinker).

52

The membrane was prehybridized at 45 °C for 2 h in 15 ml prehybridization
solution made using DIG Easy Hyb Granules (DIG High Prime DNA Labeling and
Detection Starter Kit 11; Roche Applied Science, Germany). A DIG-labeled 364 bp
fragment of the NPTII gene was created using a PCR DIG Probe Synthesis Kit (Roche
Applied Science, Germany) according to the manufacturer’s instructions. Primers used
for probe synthesis were DIGNPTII For: 5’-TGCTCCTGCCGAGGAACTAT-3’ and
DIGNPTII Rev: 5’-AATATCACGGGTAGCCAACG-3’. PCR reaction conditions
consisted of 40 cycles of 94 °C, 30 s; 57 °C, 1 min; 72 °C, 2.5 min; plus a final extension
of 72 °C, 10 min using 10 pg of pSPUD74 plasmid as template.

Twelve microliters of the labeled probe was mixed with 6 ml DIG Easy Hyb
solution and hybridized to the membrane at 45 °C overnight. Detection of the probe was
performed using a DIG High Prime DNA Labeling and Detection Starter Kit 11 (Roche
Applied Science, Germany) according to the manufacturer’s instructions using a CDP-
Star substrate and was visualized by autoradiography.

Gene expression analysis by RT-PCR

Transgenic and wild-type control plants were the same as those used for
electrolyte leakage analysis. RNA from lines containing constitutive promoters was
collected from plants growing at 22 °C. RNA from lines with cold-inducible promoters
was harvested from warm grown plants and following various exposures to 3 0C (15 min,
2 h, 24 h and following the ramp-down acclimation regime). Approximately 100 mg of
plant tissue from upper leaves of the same size as those sampled for electrolyte leakage
assay were harvested directly into liquid nitrogen and ground to a fine powder. Total

RNA was extracted using RNeasy Plant Mini Kits (Qiagen Inc., Valencia CA, USA) and

53

genomic DNA was removed by on-column DNase treatment with 3 RN ase-free DNase
set (Qiagen Inc.). Two micrograms of total RNA were reverse transcribed using
Superscript II (Invitrogen Corp., Carlsbad CA, USA) with oligo d(T)12-13 primers. PCR
reactions were carried out using 2 u] of the cDNA product as template and gene specific
primers for AtCBF] , AtCBF 3 and LeCBF] . Actin primers designed for the Tom5] gene
(GenBank accession no. U60481) from tomato were used as a loading control (Jones et
a]. 2005). Table 3.2 lists reaction conditions and primer sequences. Control reactions
were carried out using only the extracted RNA as template to verify that genomic DNA
contamination was not present following extraction.
Assay of detached leaf freezing tolerance
Plant growth of transgenic Petunia hybrida lines

Seeds from transgenic lines (Table 3.3), empty vector control, and wild type were
surface sterilized by soaking for 10 to 15 min in a 50% bleach, 0.1% Triton-X-IOO
solution with gentle agitation. Seeds were rinsed with sterile deionized water 3 to 4 times
to remove all traces of bleach and then suspended in a 0.1% sterile agar solution to
facilitate pipetting. Seeds from transgenic lines were pipetted onto 100 x 15 mm
disposable Petri plates containing Gamborg’s B5 medium (minus sucrose, kinetin and
2,4-D; plus 0.7% agar) supplemented with 100 ug/ml kanamycin. Seeds from wild-type
control plants were pipetted onto identical plates without the addition of kanamycin.
Seeds were plated at a density of ~60 seeds per plate. Plates were placed in a 22 °C, 16 h
photoperiod (100-130 umol m'2 5'1 light) chamber for three weeks. After three weeks,
kanamycin resistant transgenic plants and wild type control plants were transplanted to

50-cell trays containing 70% peat moss, 21% perlite, 9% vermiculite (Suremix, Michigan

54

Grower Products Inc., Galesburg MI, USA). Trays were placed in a 22 °C, 16 h
photoperiod (100-130 umol tn2 5'1 light) chamber and covered with humidity domes for
several days. Three weeks after transplant, plants for each line were divided into two
groups; a non-acclimated group remaining at 22 °C for an additional week before testing,
and an acclimated group subjected to a gradual cooling process. The acclimated group
was moved to 15 °C, 9 h photoperiod (100-130 umol m'2 5'1 light) for 1 week followed by
10 °C, 9 h photoperiod (100-130 umol rn'2 s'I light) for 1 week, and finally moved to 3
°C, 9 h photoperiod (30 umol m'2 5'1 light) for an additional week. At the time of testing,
non-acclimated plants were 7 weeks old and acclimated plants were 9 weeks old. Plants
in both groups were approximately the same size and had similar node number at the time
of sampling.
Electrolyte Leakage Assay

Leaf discs from the upper portion of the plant were harvested using a 0.6 cm cork
borer. The youngest leaves that could be punched to obtain complete 0.6 cm diameter
discs without cutting into the midrib were chosen for sampling. Discs were immediately
placed in deionized water and stirred gently. Approximately 120 punches were taken
from a population of 50 plants for each transgenic line. Three discs were then transferred
to each of 30 (16 x 100 mm) borosilicate culture tubes and placed on ice. When all discs
had been transferred, tubes were placed in a -1 °C controlled temperature antifreeze bath
(master bath) for 60 min. Three tubes of each line were left on ice as controls. After 60
min, a small amount of ice was added to each tube to nucleate extracellular ice formation.
Tubes were then plugged with foam and kept at -1 °C for an additional 60 min. After 60

min, three tubes of each line were moved to a second antifreeze bath at -1 °C, kept there

55

for 40 min, and removed to ice. Meanwhile, the temperature of the master bath was
lowered to -2 °C. After 20 min, three tubes of each line were moved from the master bath
to another antifreeze bath set at -2 °C for 40 min and then removed to ice (a total of 60
min at the test temperature). This process continued for all temperatures tested (generally
-1 °C to -9 °C; except -1 °C to -14 °C for acclimated pMPS]3-7 lines). Tubes were then
placed in racks on top of ice and kept at 2.5 °C to thaw slowly overnight.

The following day, 6 ml of deionized water were added to each tube followed by
shaking for three hours at room temperature to allow released electrolytes to dissolve.
The water was then transferred to a new culture tube and electrical conductiVity (L1) was
measured using a CON110 conductivity meter (Oakton Instruments, Vernon Hills IL,
USA). Plant discs remained in the original tube and were frozen to -80°C overnight to
release all electrolytes. The next day, the water from the prevrously measured tubes was
poured back into the corresponding plant disc tube, followed by shaking for three hours
at room temperature. Conductivity of the water was measured again to obtain the reading
for total electrolytes leakage (L2). Percent of total electrolyte leakage at each test

temperature was calculated by (L1+ L2)* 100. Data analysis was carried out using
Sigmaplot (SPSS Inc., USA) and SAS (SAS Institute Inc., USA) software. A sigmoidal
curve was fitted to the leakage data for each species according to the equation:

y = a1 + (a2 + (1 + exp(a3 — a4 * T ))), where y is the average percent electrolyte leakage
of the three tubes at each temperature T, using the curve fitting function of Sigmaplot.
The initial parameters were specified as a1=0.1, a2=99.9, a3=0.1 and a4=0.1 with
constraints imposed such that a1>0 and 0<a2<100. The temperature at which 50% of

total electrolytes were released (ELSO) was calculated by solving the equation EL50 =

56

((a3 — log((a2 + (50 — a1))—1))+ a4) where the parameters a] , a2, a3 and a4 are given by

the equation of the fitted curve. A total of at least two ELso values were calculated for
each transgenic line from two separate replications of the experiment. ANOVA analysis
and mean separation using Fisher’s LSD (a=0.05) were conducted on the EL50 values
using PROC GLM in SAS.
Characterization of endogenous CBF pathway
Expression of putative endogenous petCBF transcription factors

RT-PCR was used to examine expression of putative endogenous CBF genes in
Petunia hybrida ‘Mitchell.’ Four putative petCBF transcription factors have been
identified in P. hybrida ‘V26’ by Goldman et al. (2007) whose expression is induced in
response to cold (petCBFI-4) as well as drought and salinity (petCBF 4 only). Sequences
from these genes were used to design primers to examine expression patterns in P.
hybrida ‘Mitchell’. Seeds of P. hybrida ‘Mitchell’ were surface planted onto 288-cell
plug trays filled with 70% peat moss, 21% perlite, 9% vermiculite (Suremix, Michigan
Grower Products Inc., Galesburg MI, USA) and placed under intermittent mist irrigation
until at least two true leaves had developed. Plants were then transferred to a greenhouse
at 20 °C with a 16 h photoperiod until plugs were pullable (root ball holds together upon
removal from plug tray). Seedlings were transplanted to 5.7 x 5.7 x 7.6 cm pots and
grown at 20 °C, 16 h photoperiod for 4 weeks. Plants were then transplanted to 15.2 cm
pots and moved to a controlled environment growth chamber at 22 °C, 16 h photoperiod
(100-130 umol m"2 3'1 light) for 1 week prior to the beginning of cold treatments. RNA
was extracted and reverse transcribed as described above from plants grown at 22 °C and

from plants subjected to chilling at 3 °C for various periods of time (15 min, 2 h, or 24 h).

57

Two micrograms of the cDNA product was used as template in subsequent PCR analysis.
Primer sequences and reaction conditions are listed in Table 3.4.

This experiment was repeated using 7 week-old plants. Seeds from the WT3 -2
line (a wild type P. hybrida ‘Mitchell’ control line taken through tissue culture),
pSPUD74-24B-5, and pSPUD74-24B-13 were grown as described above for detached
leaf freezing tolerance assays. Seven weeks post-planting, plants were subjected to the
same chilling procedure and RT-PCR was performed as described for the first replication.
Downstream components of the CBF pathway

RT-PCR was performed to assess the impact of CBF overexpression on putative
downstream components of the CBF pathway in Petunia. A BLAST search was
conducted against the TIGR Petunia Gene Index (http://compbio.dfci.harvardedu/tgiO
using sequences for known CBF-responsive genes from other Solanaceous plants. Using
the lowest expect values (likelihood of finding such an alignment by chance alone) as
criteria, four tentative consensus sequences were selected for further investigation (Table
3.5). These sequences were used to develop gene specific primers (Table 3.6) for semi-
quantitative RT-PCR reactions aimed at determining the cold-responsiveness and CBF-
responsiveness of these sequences. Initial reactions were carried out on cDNA template
derived from warm grown (22 °C) and cold treated (3 °C for 15 min, 2 h, and 24 h) P.
hybrida ‘Mitchell’. If it appeared that a particular sequence was upregulated in response
to cold treatment, then additional reactions were carried out using cDNA template from
nonacclimated constitutive CBF-expressing transgenic lines. All reactions were
normalized by running control reactions with actin primers designed to the tomato Tom51

gene (Jones et al. 2005).

58

Identification of putative endogenous CBF transcription factors in Petunia spp.

Southern hybridization was performed to determine whether endogenous CBF
transcription factors may be present in selected Petunia species. DNA from P. hybrida
‘Mitchell’, P. axillaris (USDA 28546 and USDA 28548), P. exserta, and P. integrifolia
was extracted using a CTAB extraction method as previously described. Total DNA was
digested with either PstI or EcoRI (New England Biolabs, Beverly MA, USA) restriction
enzymes at 37 °C according to manufacturer’s instructions. Digested DNA (15 ug) was
separated and affixed to a nylon membrane as previously described.

Nucleotide sequences for known CBF genes from Solanaceous species including
LeCBF] from L. esculentum (Genbank accession no. AY497899), the CBF3-CBF1-CBF2
clusters from S. tuberosum (Genbank EU3653 84) and S. commersonii (Genbank
accession no. EU365383) and petCBF1-4 from P. hybrida ‘V26’ (Goldman et al. 2007)
were aligned using the Alian function of Vector NTI Advance 10 (Invitrogen Corp.,
Carlsbad CA, USA). A region of high conservation was selected including 65
nucleotides encoding 37% of the AP2 domain, 15 nucleotides encoding the DSAWR
signature sequence plus an additional 83 nucleotides of the sequence flanking the 3’ end.
Primers were designed to be specific to the petCBF] sequence for this region. Primer
sequences are: Forward-5’ CGGCTAGAGCTCATGACGTG 3’; Reverse-5’
GACTCCAAAGGCCTAAAAGC 3’. An initial PCR reaction using P. hybrida
‘Mitchell’ genomic DNA as template was carried out with these primers and conditions
consisting of 40 cycles of 94 °C, 30 s; 57 °C, 1 min; 72 °C, 2.5 min; plus a final extension
of 72 °C, 10min. This PCR product was separated on a 1% agarose gel and extracted

using a Gel Extraction kit (Qiagen Inc., Valencia CA, USA) Using 30 ng of gel purified

59

PCR product as template, a DIG-labeled probe was created using a PCR DIG Probe
Synthesis Kit (Roche Applied Science, Germany) according to the manufacturer’s
instructions and using the same reaction conditions as above. Southern hybridization was

performed as previously described.

RESULTS

Development of LeCBF] constitutive expression lines

Three unique single-insertion events of the 35S:.'LeCBF1 construct were
recovered from tissue culture and taken to the T1 generation. Progeny testing of T2 seed
for segregation on 100 ug/ml kanamycin plates found homozygous T1 individuals
(pXINl-17B-20 and BpXIN1-120-2) representing two of the events. No homozygous T1
plants were found for the BpXIN1-110 line, so seed from a hemizygous T 1 individual
(BpXINl-110-5) was used for subsequent tests. In order to get more lines for testing,
seed from a double-insertion T1 plant (pXIN1-25B-12) was also included in subsequent
tests.
Development of AtCBF 3 constitutively expressing lines

One single-insertion event of the 35S: :AtCBF 3 construct (pMPSl3-7) was
recovered from tissue culture and taken to the T1 generation. Progeny testing of T1
individuals for segregation on 100 ul/ml kanamycin plates failed to find any homozygous
plants. Therefore, a homogenous population was created by crossing a hemizygous T1
individual (pMPS]3-7-l2) to a wild type parent. In subsequent tests, growth of the seeds
from this cross on 100 ug/ml kanamycin plates resulted in a population consisting of only

hemizygous individuals. In order to have more lines for testing, T2 seed from a triple-

60

insertion Tl individual (pMPSl3-10-8) and T1 seed from a double-insertion To plant
(BpMPS13-101) was also included. Growth of all seeds on kanamycin assured that the
transgenic construct was present in all individuals used for testing.
Development of cold inducible AtC BF 1 lines
pSPUD74

No single insertions and only one unique double-insertion of the
AtCor15assAtCBF] construct were recovered from tissue culture. Southern analysis in
the T1 generation revealed one individual (pSPUD74-24B-13) in which a single insertion
had segregated away. Progeny testing of T2 seed showed that this plant was hemizygous
for the insertion. T2 seed from pSPUD74-24B-13 as well as T2 seed from a double
insertion T I (pSPUD74-24B-5) was used in subsequent testing.
pSPUD 73

Only two transgenic plants containing the AtC or 78: :AtCBF 1 construct were
recovered from tissue culture and each line had three insertions. Neither the T1 nor T2
seed from either of these lines showed any kanamycin resistance even at concentrations
as low a 25mg/L, making them very difficult to work with. Therefore, no plants from
this construct were used in later experiments.
CBF expression in transgenic lines used in experiments

Transgene expression level varied among the recovered constitutive expression
lines (Fig. 3.2). In the hemizygous pMPS]3-7-12 line, expression is higher compared to
all of the other AtCBF 3 over-expressing lines. Expression in the pMPS]3-10-8 and
BpMPS13-101 lines is much lower although there is clear evidence of expression in these

lines. Of the LeCBF] over-expressing lines, pXIN1-17B-20 has the highest expression

61

level, but the other lines also appear to be transcribing the transgene at moderate levels.
Expression of A tCBF 1 was not detected in either of the transgenic lines containing
A tCBF 1 under control of the cold-inducible C or] 5a promoter with or without cold
exposure (Fig. 3.3).
Basal freezing tolerance of one line is increased by constitutive expression of AtCBF 3
Basal freezing tolerance was increased in one AtCBF 3 over-expressing line,
pMPS]3-7, to a level similar to acclimated wild type plants (Fig. 3.4 and 3.5). The
nonacclimated EL50 temperature for the pMPS]3-7 line was -4.3°C, while wild type
plants had a basal ELSO = -1.7°C (Fig. 3.5), a significant increase compared to
nonacclimated wild type plants (Table 3.7A and 3.8), and not significantly different from
the ELSO temperature for acclimated wild type plants of -5.8°C. The basal freezing
tolerance of pMPS]3-10 and BpMPS13-101 did not differ from wild type. The
acclimated EL50 temperature for controls and all pMPS 1 3 transgenic lines was also not
statistically different (Table 3.7B).
Constitutive expression of LeCBF] does not affect Petunia freezing tolerance
Comparison of all LeCBF] expressing lines with wild type and empty vector
controls indicates that expression of LeCBF] did not confer a significant increase in basal
freezing tolerance as tested by electrolyte leakage assay (Fig. 3.6 and Fig. 3.7, Table
3.9A). As with the AtCBF 3 expressing lines, the EL50 value of the acclimated plants did
not differ from the controls (Table 3.93).
Petunia appears to encode endogenous CBF transcription factors
Examination of amino acid sequences for four putative Petunia CBF transcription

factors (petCBFI-4) from Goldman et al. (2007) reveals that the

62

PKK/RPAGRxKFxETRHP and DSAWR CBF signature sequences (Jaglo et al. 2001) are
largely conserved (Fig. 3.8). Expression of these petCBF genes was induced in wild type
plants in response to cold treatment at 3 °C (Fig. 3.9). Transcripts for petCBF 3 and
petCBF 4 began to accumulate after just 15 min, and after 2 h all four petCBF genes were
expressed at higher than basal levels. The cold induction of petCBF 2 expression appears
to be less than that of the other three genes. After 24 h at cold temperatures, the
expression of these genes began to taper off but levels of petCBF 1 , 3, and 4 remained
elevated compared to basal levels. After 24 h of chilling, the expression of petCBF 2 was
similar to basal levels. A similar expression pattern was seen in the pSPUD74 transgenic
lines (Fig. 3.9).

Southern hybridization using a probe created from a highly conserved region of
the already identified petCBFlin P. hybrida ‘Mitchell’ revealed at least three, and
potentially more, cross-hybridizing DNA regions within each of the other wild Petunia
species (Fig. 3.10). This suggests the presence of multiple endogenous CBF transcription
factors in each of these species.

Downstream components of the CBF cold-response pathway are conserved in Petunia

Among the CBF-responsive genes in S. tuberosum and S. commersonii is the
dehydrin-like gene, DHNlO, which is upregulated in response to both cold exposure and
CBF over-expression (Pino et al. 2008). A homolog, TC116174, is also upregulated in S.
lycopersicum in response to CBF over-expression (Zhang et al. 2004). A putative
homolog to DHNlO identified in Petunia, TC2800 (hereafter PhDHN] 0), is responsive to
cold exposure at 3 °C (Fig. 3.] 1A) as well as LeCBF] and AtCBF 3 over-expression (Fig.

3.11B). In fact, there appears to be a correlation between the level of heterologous CBF

63

expression and the expression level of PhDHN] 0. The highest AtCBF 3 expressing line,
pMPS]3-7, shows the highest level of expression of PhDHNI 0 in nonacclimated plants
(Fig. 3.11B). Likewise, in the LeCBF] expressing lines, pXIN1-17B-20 has the highest
level of expression of the transgene and has the highest expression of PhDHN] 0 in the
absence of cold treatment (Fig. 3.11B). Another putative dehydrin-like tentative
sequence from petunia, TC2907 (hereafter PhDHNZ), which is similar to DHN2 from S.
commersonii (GenBank accession no. AF3 86075), shows similar expression patterns as
PhDHNIO (Fig. 3.11). TC1671 was identified as a putative petunia homolog to a CBF-
responsive (o9 stearoyl ACP desaturase from S. commersonii (GenBank accession no.
X92847). Analysis of this tentative consensus sequence in petunia shows that it is
slightly induced in response to cold temperatures (Fig. 3.11A). However, RT-PCR
analysis failed to show that this sequence was upregulated in any of the transgenic lines
(Fig. 3.11B). A putative homolog for a CBF-responsive proteinase inhibitor, TC3472,
was identified in the TIGR petunia database by performing a BLAST search with
sequences for known CBF-responsive proteinase inhibitors found in tomato (Zhang et a].
2004), potato and S. commersonii (Pino et al. 2008). Analysis of the expression of this
TC sequence in petunia failed to show that it was upregulated in response to chilling

temperatures of 3 °C at any of the time points tested (data not shown).

DISCUSSION AND CONCLUSIONS
The CBF cold-response pathway is conserved in a number of species, including
members of the Solanaceae family such as S. lycopersicum (Zhang eta]. 2004), S.

tuberosum, and S. commersonii (Pino et al. 2007 and 2008). Our experiments show that

64

a CBF regulon is likely conserved in Petunia hybrida as well. At least four, and
potentially more, CBF transcription factors appear to be present and are upregulated in
response to a cold treatment in P. hybrida ‘Mitchell’ (Fig. 3.9). Even after 15 min at 3
°C, transcription of these endogenous CBF genes begins to increase and remains elevated
for at least 2 h (petCBF 2) or 24 h (petCBFI, 3 and 4). This is similar to the expression
pattern of CBF in Arabidopsis where exposure to temperatures of 2.5 °C induces
expression within 15 min, peak expression at 2 h, and elevated expression for at least 24
h (Gilmour et al. 1998).

Additional support for the presence of a functional petunia CBF pathway comes
from evidence that at least two downstream genes are activated by cold exposure as well
as CBF over-expression (Fig. 3.11). The dehydrin genes, DHN2 and DHN10, have been
shown to be both cold and CBF-responsive in other Solanaceous species (Pino et a]. 2008
and Zhang et al. 2004), and we have shown that putative petunia dehydrins, PhDHN2 and
PhDHN] 0, are upregulated in response to chilling. These genes are noticeably
upregulated after 24 hours of exposure to 3 °C and levels remain elevated even after the
three week rampdown acclimation regime (Fig. 3.11A). These genes are also
upregulated in non-chilled transgenic lines constitutively expressing AtCBF 3 or LeCBF]
(Fig. 3.11B). The expression level of these putative downstream genes correlates to the
expression level of the heterologous CBF genes in the transgenic lines (Fig. 3.2 and Fig.
3.11B), as transgenic lines with the highest transgene expression also show the highest
expression of PhDHN2 (TC2907) and PhDHN] 0 (TC2800).

The pMPS]3-7 transgenic line provides further evidence for a functional CBF

pathway in petunia. This line constitutively expresses AtCBF 3 at very high levels and

65

 

also shows increased basal freezing tolerance. Wild type basal freezing tolerance is near
EL50= -2.0 °C, while the pMPS]3-7 transgenic line shows basal freezing tolerance of
EL50= —4.3 °C. The basal freezing tolerance of pMPS]3-7 is not statistically different
from the freezing tolerance of fully acclimated wild type plants. Furthermore, the
pMPS]3-7 plants do not significantly increase in freezing tolerance when exposed to
acclimating conditions. Therefore, endogenous acclimation mechanisms are not capable
of increasing freezing tolerance beyond the point to which CBF over-expression does.
This does not necessarily indicate that the CBF regulon controls all of the acclimation
processes in Petunia. There may be other pathways present which fiinction in the
acclimation process, but it appears that effects of the CBF pathway and any other
potential pathways are not additive.

There appears to be a dosage-dependent effect of AtCBF 3 expression level on
freezing tolerance in transgenic P. hybrida. Two other AtCBF 3 expressing lines
(pMPS]3-10 and BpMPS13-101) express the transgene, but at levels much lower than
pMPS]3-7, and these lines show no increase in freezing tolerance. It seems that there is a
level of CBF expression that must be reached before increased freezing tolerance is
conferred. Unfortunately, there also seems to be a fitness cost to constitutively
expressing this transgene at such high levels. The pMPS]3-7 line is severely stunted and
no homozygous plants reached maturity and set seed. This line is only able to be
maintained in the hemizygous state. This severe stunting associated with high levels of
CBF expression may hinder recovery of highly-expressing lines from tissue culture.
Cells expressing CBF at high levels are probably less likely to produce healthy plantlets

and therefore recovery of high expressing lines is difficult.

66

Negative phenotypic effects of CBF expression, such as stunting and delayed
flowering, have previously been reported in Arabidopsis (Kasuga et al. 1999; Gilmour et
a]. 2000) and S. tuberosum (Pino et al. 2007). In both of these species, the use of a cold
inducible promoter minimized these negative effects (Kasuga et al. 1999; Pino et a].
2007). We attempted to employ the Cor15a cold-inducible promoter to drive expression
of AtCBF ] but were unable to verify expression of the transgene following exposure to 3
°C for up to 24 h (Fig. 3.3). In Arabidopsis, transcripts for Cor] 5a begin to accumulate
after 4 h of exposure to 2.5 °C (Gilmour et al. 1998). Cor] 5a is itself a CBF-responsive
gene (J aglo-Ottosen et al. 1998) and the successful use of a Cor15a promoter would
require the presence of native CBF transcription factors in petunia to activate it in
response to cold temperatures. We have shown that putative petCBF transcription factors
are expressed in the pSPUD74 lines (Fig. 3.9), yet the AtCor15a::AtCBF1 transgene is
not induced during cold treatment. The petCBF transcription factors may have diverged
enough from those in Arabidopsis that they no longer bind the CRT/DRE elements within
the AtCorI5a promoter. Alternatively, transcription of AtCBF 1 may be inhibited by
mutations within our transgenic construct. There is also the possibility that additional
transcription factors not present in petunia are required for activation of the C or] 5a
promoter. Creation of a double transgenic line by crossing the constitutive expression
pMPS 1 3-7 line to our pSPUD74-24B lines would reveal whether the lack of AtCBF 1
transcription is due to mutations within the construct or the lack of promoter activation by
endogenous CBF transcription factors.

It is interesting that heterologous expression of LeCBF] from tomato did not

impact freezing tolerance while high expression levels of AtCBF 3 did. Tomato belongs

67

to the same botanical family as petunia and would be expected to have genes that are
more closely related. This is indeed the case at the nucleic acid level, where the petCBF
genes identified by Goldman et al. (2007) cluster closer to the LeCBF] gene sequence
than those of the AtCBF genes (Fig. 3.12). Over-expression of LeCBF] in Arabidopsis
confers an increase in freezing tolerance, yet expression of LeCBF] or AtCBF] in tomato
has no impact freezing tolerance (Zhang et al. 2004). It was concluded that CBF over-
expression does not noticeably impact tomato freezing tolerance because the downstream
portion of the CBF regulon in tomato is much smaller than in Arabidopsis with only four
genes being upregulated in response to CBF over-expression (Zhang et al. 2004).
However, this reasoning cannot explain why LeCBF] does not change petunia freezing
tolerance while AtCBF 3 does. If AtCBF 3 expression increases freezing tolerance, then
enough of the downstream components must be present in petunia for cold. acclimation.
So why does LeCBF] not have the same effect? Again, there are several possibilities. If
there is indeed a threshold of expression that must be reached before we see increased
cold tolerance, then perhaps none of our LeCBF] -expressing lines have reached this level
of expression. It is also possible that the LeCBF] protein is less stable, resulting in the
presence of fewer functional proteins despite high transcription levels. Alternatively,
there may be differences between LeCBF] and AtCBF 3 in critical regions of the gene.
LeCBF] is more similar to petCBF as a whole, but there may be specific regions where
LeCBF] is divergent enough to make it non-functional in activating downstream petunia
genes.

In conclusion, we have shown that manipulation of the CBF cold-response

pathway in P. hybrida has the potential to increase freezing tolerance. However, the

68

 

negative phenotypic changes that accompany high constitutive expression of AtCBF 3
calls into question the feasibility of this method for improving the freezing tolerance of
commercial cultivars. Further work is needed to determine whether the use of the

C or] 5a or another cold-inducible promoter might over-come this limitation.

69

Table 3.1. Primer sequences and reaction conditions used for screening putative
transgenic P. hybrida ‘Mitchell’ plants for transgene presence in the To generation.

 

 

 

 

Primer Sequence Reaction Conditions
AtCBF] F ACGAGTTGTCCGAAGAAACC
40 cycles of 94°C 305'
At BF] R TCA AA TTGAGA A ’ ’
C GCG G C TGC 57°C, 305; 72°C, 1m; plus
f l ' f 2° 4
AtCBF3 F AATGTTTGGCTCCGATTACG ma ex‘eni‘i’: ° 7 C’
AtCBF 3 R CCTCACAAACCCACTTACCG
40 cycles of 94°C, 303;
LeCBF] F TTCGAATAACCCGAAAAAGC 56°C, 1m; 72° C, 2.5m;
plus final extension of
LCCBFI R AAAAGATCGCCTCCTCATCC 72°C, 10 min
NPT II F GAGGCTATTCGGCTATGACTG 40 03’9“ 0f 94°C, 305;

NPT II R ATCGGGAGCGGCGATACCGTA

60°C, 1.5m; 72°C, 3.5m;
plus final extension of
72°C, 10 min

 

Table 3.2. Primer sequences and reaction conditions used in RT-PCR analysis of

CBF expression in transgenic P. hybrida 'Mitchell'.
Primer Sequence

AtCBF 1 F ATTCCCAACTGCTGAAATGG

Reaction Conditions
26 cycles of 94°C, 2m;

57°C, 2m; 72°C, 2m; plus
final extension of 72°C,

 

 

 

AtCBF 1 R AATCCAGGCATGCAGAAAAG 10min
30 cycles of 94°C 2m:
At BF F T TT A TTT ’ ’
C 3 GC'ITGC G AC GCTG 60° C, 2m; 72°C, 2m; plus
AtCBF3 R GGCATGTCAACATCAGCATC final exfgsgfiifi 7‘ C’
30 cycles of 94°C, 308;
LeCBF] F GCATCAGCTGCTACI l l l lGG 56°C, 1m; 720 C, 2.5m;
plus final extension of
LeCBF] R ACAAACCCACTTGCCTGAAC 72°C, 10 min
- 26 cycles of 94°C 2m:
A t F GTGTTGGACTCTGGTGATGG ’ ’
c m 60°C, 2m; 72°C, 2m; plus
, final extension of 72°C,
Actin R TCAGCAGTGGTGGTGAACAT

10 min

 

70

Table 3.3. Plant generation and transgene copy number for the transgenic P.

hybrida ‘Mitchell’ lines tested in freezing tolerance eygeriments.

 

 

Transgene
Generation and Zygosity of Copy
Line ID Tested Individuals Number
35S::AtCBF3
Mitchell 6 (WT) >< pMPS]3-7-12 all hemizygous seed from cross 1
BpMPS 1 3-101 T1 seed from hemizygous To 2
pMPS]3-10-8 T2 seed from a 3 insertion T. 3
3SS::LeCBF1
BpXIN1-110-5 T2 seed from hemizygous T1 1
BpXINl-120-2 T2 seed from homozygous T1 1
pXIN1-17B-20 T2 seed from homozygous T1 1
pXIN1-25B-12 T2 seed from 2 insertion T1 2
AtCorlSa::AtCBFl
pSPUD74-24B-5 T2 seed from 2 insertion T1 2
pSPUD74-24B-13 T2 seed from a hemizygous T1 1

 

Table 3.4. Primer sequences and reaction conditions for
RT-PCR of putative endogenous petunia CBF genes.
Reaction conditions consisted of 26 cycles of 94 °C, 30
s; 56 °C, 1 min; 72 °C, 2.5 min; plus final extension of
72 °C, 10 min.

 

Primer Sequence
petCBF] F ATTCCCAACTGCTGAAATGG
petCBF] R AATCCAGGCATGCAGAAAAG
petCBF2 F TGCTTGCTTGAACTTTGCTG
petCBF2 R GGCATGTCAACATCAGCATC
petCBF3 F GCATCAGCTGCTACTTTTTGG
petCBF3 R ACAAACCCACTTGCCTGAAC
petCBF4 F TGCAGAAATGGCAGCTAGAG
petCBF 4 R AGGAAGCCGGGATAGGTAAC

Actin F GTGTTGGACTCTGGTGATGG

Actin R TCAGCAGTGGTGGTGAACAT

 

 

71

.UocmgouD—u HO: mm; mmoco>mmcoamohummmv

83065 x MOO mm 3 030% 8:: maimeaxoego T393 Ea mkmbaw Eofi 080838 <ZM =0 mun—Fm 3 0035.800 mmocozmcoqmocimu o

E am was .5 N .EE 2 .EE 0 now Do m “a 00:20 £5300 E0¢ .0808on <Zm :0 mun—FM 3 005.5800 8000320397200

0

>0w.£:.E_:.50:.333\\a:: E0; 9 0:89:00 {fits—00.0322.BEdEanohat: E0: 0008:: Um. a

 

 

 

 

 

a» a» 2589 32:00 €150 woom s a 9.: Easess .0. £83.?
0: m0» :0 _ O... 883800 woom N: 3 0:5 2:305:80 a. mammox
n5,4 30305 08 . ..
00. _ E as 0:. 550.0 E ..
x 0: Hemp—l 03:5 H.H.; . woos w a a u 29%sz
. 00. 023:5 voom N0 E EEN $230303 .N $02 5%
m0» m0» commuk 2: ZIQV 5205 Bow. B E 0:5 55.3005 .m Rome _ Orr
00.2-5630D voom 3 E wSEN 20:23:00.0 a wig _UP
00>_m:00m2-u_m0 325030-200 access: 2035.800 oosocflom 862$ bosom an: 0:50:00
2 Oh 2:30; tench—E OF

 

mw0.0;tO «Each. _EEBOQ

 

3:00 0>_m:00m0~_ ”EU :305—

 

03.53 2.550% 5 538.9880 “EU 0.5 200 00 030%“: E 550.298 :05 0:0 macaw uogzwocimo («0 $205.5 028:5 .m.m 033—.

72

Table 3.6. Primer sequences and reaction conditions used in RT-PCR analysis of

putative downstream genes in the endogenous CBF-pathway in Petunia hybrida
'Mitchell'.

Primer Sequence Reaction Conditions

TC2800 F GGGATGCACTTCACATTGG

17 cycles of 94°C, 2m;
TC2800 R TCCTTCTTATCCTTCTTGTCCTTC 55°C, 2m; 72°C, 2m;

plus final extension of

TC3472 F ACCAGGCCATGCTAACACC 72°C, 10 min

TC3472 R GGCAGAAACCAAAAATACACC

 

TC1671 F AAGGCTGATGTCCTCCAAAG

’ 16 cycles of 94°C, 2m;
TC1671 R GCAATGACCCAAACTCCTG 55°C, 2m; 72°C, 2m;

plus final extension. of

TC2907 F CGAGTGATGAGGAGGAGGAAG 72°C, 10 min

TC2907 F GTGGTGGTGGTGGTGTTG

 

. 20 cycles of 94°C 2m“
F , ’
Actin GTGTTGGACTCTGGTGATGG 60°C, 2m; 72°C, 2m;

plus final extension of

Actin R TCAGCAGTGGTGGTGAACAT 72°C, 10 min

Table 3.7. ANOVA for effect of transgenic line on non-acclimated (A) and
acclimated (B) EL50 temperature of AtCBF 3-expressing P. hybrida ‘Mitchell’ lines.

 

 

 

 

A. Dependent variable: Sum of Mean F

EL50 temperature Source DF Squares Square Value Pr > F
Model 4 13.34 3.34 8.67 0.00
Error 11 4.23 0.38
Total 1 5 1 7.5 8

B. Dependent variable: Sum of Mean F

EL50 temperature Source DF Squares Square Value Pr > F
Model 4 4.55 1.14 1.72 0.22
Error 11 7.27 0.66

Total 15 1 1.83

73

Table 3.8. Pairwise comparisons with Fisher's LSD between non-
acclimated EL50 values of different AtCBF 3 expressing lines and
controls. Shown are p-values for each comparison.

 

 

$113: 13;]ng pMPS]3-7 pMPS]3-10
Empty Vector 0.555
pMPS]3-7 0.000 0.001
pMPS]3-10 0.637 0.924 0.001
BpMPS13-101 0.185 0.138 0.009 0.161

 

Table 3.9. ANOVA for effect of transgenic line on non-acclimated (A) and
acclimated (B) EL50 temperature of LeCBFI-expressing P. hybrida ‘Mitchell’

 

 

 

 

lines.
A. Dependent variable: Sum of Mean F
ELso temperature Source DF Squares Stglare Value Pr > F
Model 5 3.49 0.7 1.73 0.2
' Error 12 4.84 0.4
Total 17 8.33
B. Dependent variable: Sum of Mean F
ELso temperature Source DF Squares Square Value Pr,> F
Model 5 5.37 1.07 1.77 0.19
Error 12 7.28 0.61
Total 17 12.65

 

74

 

pSPUD73

 

 

 

 

 

 

 

 

 

LB[ Ubi3 Pro [ NPTII [ Ubi3 Term Hgtcma Pro AtCBF1 Term lRB
pSPUD74

LB[ Ubi3 Pro [ NPTII I Ub13 Term HAtCor15a Prol AtCBF1 ] Term IRB
pMPS13

LBL NOS Pro ] NPTII I NOS Term l—LCamvsss [ AtCBF3 Term ]RB
pX|N1

LB[ nos Pro I NPTII | nos Term HCaMV 353T LeCBFt | Term JRB

 

 

 

 

Figure 3.1. Diagram of constructs used for Agrobacterium-mediated transformation of P.
hybrida ‘Mitchell.’ All constructs contain NPT]! as a selectable marker for kanamycin
resistance. pSPUD73 contains Arabidopsis thaliana CBF '1 behind the cold-inducible
AtCor78 promoter. pSPUD74 contains Arabidopsis thaliana C BF 1 behind the cold-
inducible AtC or] 5a promoter. pMPS]3 and pXIN 1 contain Arabidopsis thaliana CBF 3
and Lycopersicon esculentum CBF], respectively, behind the strong constitutive CaMV
35S promoter.

 

Wild Type
leNl-ZSB
leN1-120

Wild Type
pMPS]3-10
pMPS]3-IO]

i
‘ LeCBF]

. leN1-17B
' pXINI-IIO

'T
M
U)
0..
E
0.

 

AtCBF3

as” M n a.» Actin “ - - “ - Actin

 

A B

 

 

 

Figure 3.2. Gene expression analysis by semi-quantitative RT-PCR of P. hybrida
‘Mitchell’ transgenic lines containing 355: :AtCBF 3 (A) or 35S::LeCBF1 (B). RNA was
isolated from plants grown at 22 °C.

75

pSPUD74-24B-5 Rampdown acclimated

pSPUD74-24B-5 24 hr 3°C

pSPUD74-24B-13 Nonacclimated
pSPUD74-24B-5 Nonacclimated
pSPUD74-24B-5 15 min 3°C
pSPUD74-24B-5 2 hr 3°C

pSPUD74-24B-13 24 hr 3°C

pSPUD74 plasmid

I
to

AtCBF1

Actinmnmuuflumw-

Figure 3.3. RT-PCR analysis of A tCBF 1 expression in pSPUD74 transgenic lines
following various exposures to cold temperatures. Nonacclimated plants were grown at
22 °C and rampdown acclimated plants were grown 7 d at 15 °C SD, 7 d at 10 °C SD, and
7 d at 3 °C SD.

76

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

wud Type 'Mitchell' A '20 Empty Vector 120
B
- 100 ., .. - 100
i ._ i .. .-
r r80 3, ., .80 3,
5‘3 ' a
"’ _l
i 40 °\° 40 ..\°
i i ~20 i h .20
. 1,1 , e 2 E II o
-10 -8 -6 -4 -2 0 -1o -8 -6 -4 -2 0
Temperature (°C) Temperature (°C)
120
pMP313'7'12 C pMPS13-10-8 D 12°
i " - 100 .- __ . . 100
T
r ' 80 § .. _ 80 a
(U
x .. x
L 60 g ! l 60 g
* 4° °\° E . 40 39
- llllll"° l l
-2 0 0

-10 -8 -6 «1 -10 -8 -6 -4 -2

 

Temperature (°C) Temperature (°C)
BpMPS13-101 E 120
,_ - 100
l t 80

O)
O
% Leakage

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ii i r20
. I III 0
-10 -8 a -4 -2 0

Temperature (°C)

Figure 3.4. Percent electrolyte leakage at each temperature tested for wild type (A),
empty vector control (B), and transgenic lines (C-E) containing the 35S: :AtCBF 3
(pMPSl3) construct. Nonacclimated plants (black bars) were grown at 22 °C LD and
acclimation was accomplished by exposing plants to 15 °C SD for 1 week, 10 °C SD for 1
week, then 3 °C SD for 1 week (grey bars). Leakage data at each temperature averaged
over 6 measurements for empty vector and transgenic lines; 24 measurements for wild

type. Standard deviation shown by error bars.

77

 

5% BpMPS13-101

 

5W pMps13-1o

r...

-%/ pMPS13-7

%%%%WWW Empty Vector

 

_‘. .
l .
-_ '

Wild Type

.‘4‘1

- Nonacclimated 7 ’
Acclimated

 

 

|
-10 -8 '5 '4 ’2 O
EL50 Temperature (°C)

 

Figure 3.5. ELso temperatures of nonacclimated and acclimated AtCBF 3 constitutively
over-expressing lines. pMPS 1 3-7 is significantly more freezing tolerant than the control
lines prior to acclimation (starred bar). Following our acclimation regime (7 d at 15 °C
SD, 7 d at 10 °C SD, and 7 d at 3 °C SD), there is no significant difference between any
of the transgenic lines and the control lines. Standard deviation shown by error bars.

78

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

wad Type 'Mitchell' A 120 Empty Vector B 120
.. I ._ .. ,_ r 100 - 100
_ T
g r 3
r 60 3 * 60 8
_l _l
+ 40 39 » 40 33
i i >20 i h »20
ti] 0 , _ I] 0
-10 -8 -6 -4 -2 0 -10 -8 45 -4 -2 0
Temperature (°C) Temperature (°C)
pXIN1-17B-20 C 120 pXIN1-258-12 D 120
« V ~ 100 r 100
-80 3, ~80 g,
3 $2
»60 3 ~60 g
_l _l
r 40 '3‘3 r 40 Be
» 20 i i 20
w .0 , l l ,l 0
-10 -8 -6 -4 -2 0 -1O -8 -6 -4 -2 0
Temperature (°C) Temperature (°C)
120
BpXIN1-110-5 E BpXIN1-120-2 F 12°
T ~ 100 ~ 100
I It I; m r i~ I
: ; . 80 a) r 80 8,
g T 3 g g
~60 g W -60 3
_J _l
b 40 o\° - 40 53
ii - 20 i L -20
_l ._.. LIL L , W l _ 0 ‘ r, i l 7 [I o
-10 -8 -6 -4 -2 0 -1O -8 -6 -4 -2 0

Temperature (°C)

Temperature (°C)

Figure 3.6. Percent electrolyte leakage at each temperature tested for wild type (A),
empty vector control (B), and transgenic lines (C-F) containing the 35S::LeCBF1
(pXIN 1) construct. Nonacclimated plants (black bars) were grown at 22 °C LD and

acclimation was accomplished by exposing plants to 15 °C SD for 1 week, 10 °C SD for 1

week, then 3 °C SD for 1 week (grey bars). Leakage data at each temperature averaged
over 6 measurements for empty vector and transgenic lines; 24 measurements for wild

type. Standard deviation shown by error bars.

79

 

%5€///////////////////////////////////////////// BpXIN1-120

m
. 7'), .4

L27/////////////////////////////////////% Bpxm1-11o

T'S‘fii'ififi j.“ :r i

      

u. . 5
. ,. .u‘
c‘,_..' _

57////////////////////////////////////////////////////////////% pXIN1-25B

m ‘1»:

'5’."////////////////////////////////////////////////////////, leN1-17B

a.
an

 

7.x, .~
~_:- {J

WWW Empty Vector

'1
g .

. 54
,1 f..- _' -. .{J

 

 

W Wild Type

__ ”.n 3374‘

 

- Nonacclimated
’///.// Acclimated

 

 

I ' ' I

 

~10 -8 -6 -4 -2 O
EL50 Temperature (°C)
Figure 3.7. EL50 temperatures of nonacclimated and acclimated LeCBF] constitutively
over-expressing lines. Nonacclimated plants were grown at 22 °C LD and acclimation
was accomplished by exposing plants to 15 °C SD for 1 week, 10 °C SD for 1 week, then
3 °C SD for 1 week. There are no significant differences between any of the transgenic
lines and the control lines.

80

l 50

 

 

 

 

 

 

 

 

 

petCBFl MDIFGSYYSDTLPAASAPTFWPLDVPEYSSPISDNSSCSNNRANHSDEEV
petCBF2 MDIFGSYYSDIL ————————— PIELPEYSSPMSDNSSCSNYRANHSDDEV
petCBF3 MDIFARYYSDQLPIASAATFWPLEVAEYSSPMSD---ISNNRANLSDEEV
petCBF4 MDIFGRYYSDQLPIASAATFWPLEVAEYSDNSSS---SSNNRANVSDEEV
AtCBFl -------------------------- MNSFSAFSEMFGSDYEPQGGDYCP
51 g 100
petCBFl MLASN PKKRAG E VYRGVRKRNSGKWVCEVREPNKQSRIWL
petCBF2 MLASN PKKCAG F VYRGVRKRN-GKWVCEVREPNKKSRIWL
petCBF3 MLASN pxxaae r VYRGVRKRSSGKWVCEVREPNKKSRIWL
petCBF4 MLASN PKKRAG r VYRGVRKRNSGKWVCEVREPNKKSRIWL
AtCBFl TLATS PKK as E IYRGVRQRNSGKWVSEVREPNKKTRIWL
101 150
petCBFl GTFPTAEMAARAHDVAAIAFRGRSACLNFA :LPTPASSDPKDIQKA
petCBF2 GSFPTAEMAARAHDVAAIALRGRSACLNFA _LPIPASSNPKDIQKA
petCBF3 GTYITAEMAARAHDVAAIALRGRSACLNFP JLHIPASSKAKDIQKA
petCBF4 GTYSTAEMAARAHDVAAIALRGRAACLNFP fLPIPASSKAKDIQKA
AtCBFl GTFQTAEMAARAHDVAALALRGRSACLNFP _LRIPESTCAKDIQKA
151 200
petCBFl AAEAAEAFRPLESEGVHSAGEESKEESTTPETAES ————— MYFMDEEALF
petCBF2 AAEAAKAFR -------- ESGEESKEESSTRETPEK ————— MFFMDEEALF
petCBF3 ATEAASAFQ ------------ ESKEEGTTPETPEK ----- MLFMDEEALF
petCBF4 ATEAAATAF ———————————— LEPGE---PETRKKN——-—MLFMDEEALF
AtCBFl AAEAALAFQDETCDTTTTNHGLDMEETMVEAIYTPEQSEGAFYMDEETMF
201 242
petCBFl CMPGLLANMAEGLMLPPP-QCSEVGDHFMEADADMPLWSYSV
petCBF2 CMPELLANMAEGLMLPPPSQCSDVGEHFMDADVDMPLWSYSI
petCBF3 YMPGLLANMAEGLMLPLPPQCSEVGDHFMEAAADMPLWSYSF
petCBF4 CMPGLLANMAEGLMLTPP---QCYGEHFMEADAEVPLWSY-—
AtCBFl GMPTLLDNMAEGMLLPPP—SVQWNHNYDGEGDGDVSLWSY--

Figure 3.8. Alignment of petCBF amino acid sequences obtained from Goldman et al.
(2007) and AtCBF] sequence (Pubmed Gene ID: 828653). Shown in boxes is the
matching of the petCBF sequences with the “CBF signature sequences”,

PKK/RPAGRxKFxETRHP and DSAWR (J aglo et al. 2001). Grey shading denotes
where the petCBF amino acids differ from the signature sequence.

81

 

'O

93

(U

'o E

.03. ‘0 "5

m 93 U 0

E? g *3 m

:(r) O o C
859:0'fig’ég, o?
wgmaom-o' 03,8

C bcwho.E° D.

“C Clo-c C-CCUEmE
a) Zs—Naovc h E
*5 ._._._._ZNOu)-C§(U
2% 0555-5779999“?
8=§$=§§§eeeee$e
EEEE-Eg-E-Ewwwww‘rw
o comma: V
29~§EEEE§EEEEBE
ggggigfignasnsDa
cascaaaaaaa
EEEEQEIQQQQQQQQ

petCBF1 -c— - a. -
petCBF2 in ;;;—i_ ‘

   

petCBF3 C- C W .- -

'

petCBF4 . 0 ~ .1. D a .

Actin -n---m~-m~--~-~

 

 

 

Figure 3.9. RT-PCR analysis for expression of petCBF1 -4 in wild type P. hybrida
‘Mitchell’ and pSPUD74 transgenic lines following chilling at 3 °C for various time
periods. Nonacclimated plants were grown at 22 °C and rampdown acclimated plants
were grown at 15 °C SD for 1 week, 10 °C SD for 1 week, then 3 °C SD for 1 week.
WT3-2 is a wild type line recovered from tissue culture and P. hybrida ‘Mitchell’ is a
wild type that has not undergone tissue culture.

82

>
'03

 

T: %

_ccoco _Ccoco
gas ere
Ema: %.§NN Lg
m.W.‘0mHL€U.W.‘0<Uk
fottttjottt':
ELULUCDOELULBQJD:
iiifigfiii‘fig
ammo-Scmmm-S
QOLQQLDLCL'L‘L'CL'QLD;

i
a 'h ‘1
m...

 

Figure 3.10. Southern hybridization with genomic DNA from four Petunia species
digested with PST I (A) or ECO RI (B) restriction enzymes. Probe is a 163 nucleotide
fragment from a highly conserved region of petCBF1 from P. hybrida ‘Mitchell’.

83

 

 

 

 

'o
93
(0':
.3. E:
m '-='
.§ §
E o. o
v C 0 0
N 5% EOQQQQQQK
0- VUUEON..eeNNN
3:9 NQKNNNFNNNNL.
N..ONN....
ngjdgwm”°"djm‘°8%
..-.N ..-.'\‘_l ‘—
fifiwifiefifiiseegtazzs
Q-QQINQI-Q-o“‘—‘- #422
>~>~mcbm>>~g$QZZ—-€
.2: cc 2__><><
..EEE..§§QXXQGE
CLO. llaamaammm
l .
TC2907 Q. Q'- ‘Q .1...
Tczsoo - Q Q - --. .. -- an
TC1671 -- m “at“:

Actin Qfiflflfi.flfl-flflfiflflfl

 

 

 

Figure 3.11. Expression of putative downstream components of the CBF-regulon in P.
hybrida ‘Mitchell’ determined by RT-PCR. Cold-responsiveness of sequences was
verified in wild type plants (A) and CBF-responsiveness was determined in
nonacclimated transgenic lines (B). TC numbers and sequences were obtained from
http://compbio.dfci.harvard.edu/tgi/

 

 

 

 

 

 

 

 

 

 

 

 

 

 

—L AtCBFl (0.0766)
AtCBFZ (0.0686)
AtCBF3 (0.0679)
Lecrrl (0.1268)
‘ petCBF3 (0.0694)
petCBF4 (0.1268)
petCBF2 (0.1313)

 

 

Figure 3.12. Phylogenetic tree showing relationship between nucleic acid sequences for
CBF transcription factors from various species.

84

REFERENCES

Baker 3., K. Wilhelm, and M. Thomashow. 1994. The 5’-region of Arabidopsis thaliana
cor15a has cis-acting elements that confer cold-, drought- and ABA-regulated
gene expression. Plant Mol. Biol. 24:701-713.

Chen H.H. and PH. Li. 1980. Characteristics of cold acclimation and deacclimation in
tuber-bearing Solanum species. Plant Physiol. 65:1146-1148.

Cook D., S. Fowler, 0. F iehn, and M. Thomashow. 2004. A prominent role for the CBF
cold response pathway in configuring the low-temperature metabolome of
Arabidopsis. Proc. Natl. Acad. Sci. 101:15243-15248.

Fowler S. and M. Thomashow. 2002. Arabidopsis transcriptome profiling indicates that
multiple regulatory pathways are activated during cold acclimation in addition to
the CBF cold response pathway. Plant Cell 14:1675—1690.

Gilmour S., D. Zarka, E. Stockinger, M. Salazar, J. Houghton, and M. Thomashow. 1998.
Low temperature regulation of the Arabidopsis CBF family of AP2 transcriptional
activators as an early step in cold-induced COR gene expression. Plant J. 16:433-
442.

Gilmour S., A. Sebolt, M. Salazar, J. Everard, and M. Thomashow. 2000. Overexpression
of the Arabidopsis CBF 3 transcriptional activator mimics multiple biochemical
changes associated with cold acclimation. Plant Physiol. 124: 1 854-1 865.

Gilmour S., S. Fowler, and M. Thomashow. 2004. Arabidopsis transcriptional activators
CBF], CBF 2, and CBF 3 have matching functional activities. Plant Mol. Biol.
54:767-781.

Goldman S.L, S.V. Rudrabhatla, M. Parini, M. Styczynski, M.R. Raab, Univ.of Toledo,
and Agrivida Inc. Plants expressing environmental stress tolerances having
petunia CBF genes therein. Patent application WO 2007/048030. April 26, 2007.

Guy C., K. Niemi, R. Brambl. 1985. Altered gene expression during cold acclimation of
spinach. Proc. Natl. Acad. Sci. 82:3673-3677.

Hsieh T., J. Lee, P. Yang, L. Chiu, Y. Charng, Y. Wang, and M. Chan. 2002a.
Heterology expression of the Arabidopsis C -Repeat/Dehydrati0n Response
Element Binding Factor 1 gene confers elevated tolerance to chilling and
oxidative stresses in transgenic tomato. Plant Physiol. 129:1086-1094.

Hsieh T., J. Lee, Y. Charng, and M. Chan. 2002b. Tomato plants ectopically expressing

Arabidopsis CBF] show enhanced resistance to water deficit stress. Plant Physiol.
130:618-626.

85

Jaglo K., S. Kleff, K. Amundsen, X. Zhang, V. Haake, J. Zhang, T. Deits, and M.
Thomashow. 2001. Components of the Arabidopsis C-repeat/dehydration-
responsive element binding factor cold-response pathway are conserved in
Brassica napus and other plant species. Plant Physiol. 127:910-917.

J aglo-Ottosen K., S. Gilmour, D. Zarka, O. Schabenberger, and M. Thomashow. 1998.
Arabidopsis CBF] overexpression induces COR genes and enhances freezing
tolerance. Science 280: 104-106.

Kitashiba H., T. Ishizaka, K. Isuzugawa, K. Nishimura, and T. Suzuki. 2004. Expression
of a sweet cherry DREBI/CBF ortholog in Arabidopsis confers salt and freezing
tolerance. J. Plant Physiol. 161:1171-1176.

Kasuga M., Q. Liu, S. Miura, K. Yamaguchi-Shinozaki, and K. Shinozaki. 1999.
Improving plant drought, salt, and freezing tolerance by gene transfer of a single
stress-inducible transcription factor. Nature Biotechnol. 17:287-291.

Maruyama K., Y. Sakuma, M. Kasuga, Y. Ito, M. Seki, H. Goda, Y. Shimada, S.
Yoshida, K. Shinozaki, and K. Yamaguchi-Shinozaki. 2004. Identification of
cold-inducible downstream genes of the Arabidopsis DREBlA/CBF3
transcriptional factor using two microarray systems. Plant J. 38: 982-993.

Owens C., M. Thomashow, J. Hancock, and A. Iezzoni. 2002. CBF] orthologs in sour
cherry and strawberry and the heterologous expression of CBF] in strawberry. J.
Amer. Soc. Hort. Sci. 127:489-494.

Pennycooke J ., M. Jones, and C. Stushnoff. 2003. Down-regulating a-Galactosidase
enhances freezing tolerance in transgenic petunia. Plant Physiol. 133:901-909.

Pino M.T., J .S. Skinner, E.J. Park, Z. Jeknic, P.M. Hayes, M.F. Thomashow, and T.H.H.
Chen. 2007. Use of a stress inducible promoter to drive ectopic AtCBF expression

improves potato freezing tolerance while minimizing negative effects on tuber
yield. Plant Biotechnol. J. 5:591-604.

Pino M.T., J .S. Skinner, Z. Jeknic, P.M. Hayes, A.H. Soeldner, M.F. Thomashow and
T.H.H. Chen . 2008. Ectopic AtCBF! over-expression enhances freezing tolerance

and induces cold acclimation-associated physiological modifications in potato.
Plant Cell Environ. 31:393-406.

Qin F., Y. Sakuma, J. Li, Q. Liu, Y. Li, K. Shinozaki, and K. Yamaguchi-Shinozaki.
2004. Cloning and functional analysis of a novel DREBl/CBF transcription factor

involved in cold-responsive gene expression in Zea mays L. Plant Cell Physiol.
45:1042-1052.

86

Steponkus P.L., M. Uemura, R.A. Joseph, and SJ. Gilmour. 1998. Mode of action of the
COR15a gene on the freezing tolerance of Arabidopsis thaliana. Proc. Natl. Acad.
Sci. 95: 14570-14575

Stockinger E., S. Gilmour, and M. Thomashow. 1997. Arabidopsis thaliana CBF]
encodes an AP2 domain-containing transcriptional activator that binds to the C-
repeat/DRE, a cis-acting DNA regulatory element that stimulates transcription in
response to low temperature and water deficit. Proc. Natl. Acad. Sci. 94:1035-
1040.

Thomashow M. 1999. Plant cold acclimation: Freezing tolerance genes and regulatory
mechanisms. Annu. Rev. Plant Physiol. Plant Mol. Biol. 50:571-599.

Vogel J ., D. Zarka, H. Van Buskirk, S. Fowler, and M. Thomashow. 2005. Roles of the
CBF 2 and ZATIZ transcription factors in configuring the low temperature
transcriptome of Arabidopsis. Plant J. 41 :195-21 1.

Yelenosky G., and C. Guy. 1989. Freezing tolerance of citrus, spinach, and petunia leaf
tissue. Plant Physiol. 89:444-451.

Zhang X., S. Fowler, H. Cheng, Y. Lou, S. Rhee, E. Stockinger, and M. Thomashow.
2004. F reeling-sensitive tomato has a functional CBF cold response pathway, but
a CBF regulon that differs from that of freezing-tolerant Arabidopsis. Plant J.
39:905-919.

87

CHAPTER 4

ASSESSING EFFECTS OF CBF OVER-EXPRESSION ON HORTICULTURAL
TRAITS IN PE T UNIA H YBRIDA ‘MITCHELL’

ABSTRACT

Heterologous expression of CBF transcription factors has previously been shown
to effectively increase freezing tolerance of many plant species, including petunia.
However, strong constitutive CBF expression often results in negative phenotypic
changes, such as stunting and delayed flowering. Here we show that while high
constitutive expression of AtCBF 3 increases basal freezing tolerance of one transgenic
petunia line, deleterious phenotypic changes occur such as increasing days to flower by
over 30%, reduced flower size, and reduced biomass at flowering. These negative
phenotypes were either less severe or not present in other transgenic lines, but transgene
expression was lower in these lines and they showed no increase in fi'eezing tolerance. It
is concluded that constitutive expression of AtCBF 3 at levels high enough to increase

freezing tolerance brings about negative effects on horticulturally important traits.

INTRODUCTION:

Frequent night temperatures below freezing severely limit the selection of
ornamental bedding plant species that can be grown in early spring and late fall.
However, in some species, exposures to low non-freezing temperatures induce an
increase in freezing tolerance, referred to as cold acclimation. Following acclimation,

plants survive temperatures below levels that were previously lethal (Thomashow 1999).

88

Increased freezing tolerance is the result of many genetic changes (Guy et al.
1985), including those regulated by the CBF cold response pathway. This well-
characterized pathway in Arabidopsis thaliana L. consists of three functionally redundant
transcription factors, AtCBF1-3 (Gilmour et al. 2004). Downstream components of the
pathway contain a cis-acting element called the Q-gepeat/thydration gesponsive element
(CRT/DRE) to which CBF binds (Stockinger et al. 1997, Gilmour et al. 1998).
Expression ofAtCBF1-3 is induced in response to low temperatures, bringing about an
increase in expression of downstream genes, such as the COR (QM-Responsive) genes,
and ultimately leading to an increase in freezing tolerance (J aglo-Ottosen er al. 1998).

Heterologous expression of Arabidopsis CBF genes has proven to be a useful
method for increasing freezing tolerance of several plant species, including Brassica
napus L. (Jaglo et al. 2001), Fragaria Xananassa Duchesne (Owens et al. 2002) and
Solanum tubersosum L. (Pino et al. 2007; Pino et al. 2008). However, expression of
AtCBF transcription factors behind the strong constitutive cauliflower mosaic virus
(CaMV) 35S promoter often results in deleterious phenotypic effects. For instance, when
AtCBF 3 is over-expressed in Arabidopsis, stunting, delayed flowering, and prostrate
growth habits have been observed (Liu et al. 1998; Kasuga et al. 1999; Gilmour et al.
2000). Constitutive expression of AtCBF 1 -3 in S. tuberosum reduces or inhibits tuber
formation, delays flowering, and causes stunting (Pino et a1. 2007). Solanum
lycopersicum L. plants expressing either AtCBF 3 or LeCBF] (a CBF homolog from S.
lycopersicum) behind the CaMV 35S promoter also display stunted growth and delayed

flowering (Zhang et al. 2004).

89

Petunia hybrida Vilm. (petunia) is a major horticultural crop with the ability to
cold acclimate (Yelenosky and Guy 1989; Pennycooke et al. 2003). However, further
improvements to the cold tolerance of petunia would benefit both consumers and
commercial growers by allowing this species to be grown earlier in the spring and later
into the fall, when cool temperatures limit the growth of other bedding plants. Therefore,
one of two CBF transcription factors, either AtCBF 3 or LeCBF] , have been transformed
into P. hybrida ‘Mitchell’ behind the CaMV 35S promoter in an effort to further increase
the freezing tolerance of this species. One transgenic line, pMPS]3-7, expressing
AtCBF 3 at high levels has previously been shown to have increased basal freezing
tolerance compared to wild type (chapter 3). The objective of this study is to determine
whether constitutive expression of these genes has deleterious effects on traits of

horticultural importance in petunia.

MATERIALS AND METHODS

Agalvsis of horticultural traits in transgenic lines

Transgenic P. hybrida ‘Mitchell’ lines expressing AtCBF 3 or LeCBF] behind the
strong constitutive CaM V 35S promoter were created as described in chapter 3. These
lines were tested to determine the effect of constitutive CBF expression on horticultural
traits. Seeds from transgenic lines, empty vector control, and wild type were surface
sterilized and planted onto petri plates as described in chapter 3. Seeds were plated at a
density of ~1 00 seeds per plate. Plates were placed in a 22 °C, 16 h photoperiod (100-
130umol m'2 3'1 light) chamber for three weeks. After three weeks, 10 plants of each

genotype were transplanted to 10 cm round pots filled with 70% peat moss, 21% perlite,

9O

9% vermiculite (Suremix, Michigan Grower Products Inc., Galesburg MI, USA). Plants
were left in a 22 °C, 16 h photoperiod chamber with increasing light intensity for 1 week
to harden plants in preparation for transfer to a greenhouse. Plants were then transferred
to a greenhouse at 20 °C with 16 h daylight extension lighting to grow until the first

. flower opened. On the day the first flower opened, days to flower, number of fully open
flowers, flower bud number, length of flower-bearing stem, leaf number on the flower-
bearing stem, length and width of longest leaf, open flower corolla tube length and
corolla rim diameter were determined. Aboveground biomass was harvested and dried in
an oven at 70 0C for five days for dry mass determination. In the second replication,
additional measurements were made including length of the primary stem and leaf
number on this stem. Genotype effects were determined by ANOVA using PROC GLM
of SAS (SAS Institute Inc., USA) and mean separations were carried out using two-tailed
Dunnett comparisons with wild type as the control. Two replications of this experiment

were conducted.

RESULTS
CBF over-expression impacts horticultural traits
Constitutive CBF expression impacted several traits of horticultural importance
compared to wild type (Tables 4.1 and 4.2). There was a significant effect of replication,
so data from the two replications were analyzed independently. Time to flowering of the
pMPS 1 3-7 transgenic line was increased compared to wild type. In the first replication,
wild type plants flowered an average of 70 d afier planting, compared to 94 d for

pMPS13-7 (Table 4.1). Days to flowering of all other transgenic lines in the first

91

replication did not differ from wild type. In the second replication, wild type plants took
77 d to flower while pMPS]3-7 plants flowered 127 d after planting (Table 4.2).

The pMPS 1 3-7 line flowered exclusively on lateral branches in both replications,
while the first open flower for wild type and other transgenic lines occurred on the
primary stem more often than on the lateral branches. In the first replication, length of
the first flower-bearing stem was shorter than wild type for pMPS]3-7 and pXIN1-25 B,
but the number of leaves on the flower-bearing stems did not differ (Table 4.1),
indicating reduced intemode length. In the second replication, pMPS]3-10, BpMPSl3-
101, pXIN1-25B, and BpXIN1-110 all had shorter flower-bearing stems than wild type,
but leaf number did not differ (Table 4.2). Leaf numbers on the primary stems at time of
flowering were determined in the second replication (Table 4.2) and indicate that
pMPS 1 3-7 plants flowered at approximately the same physiological time as wild type,
and the delayed flowering is the result of a reduced development rate,

Open flower number on the first day of flowering was not different from wild
type in any lines (Tables 4.1 and 4.2). Flower bud number was decreased in pMPS]3-7
during the first replication compared to wild type, with wild type plants having an
average of 28 unopened visible flower buds on the day of flowering, compared to only 17
on pMPS]3-7 (Table 4.1). In the second replication, pMPS]3-7 showed a slight increase
in visible buds on the day of flowering, having 31 per plant compared to 24 for wild type
(Table 4.2).

Length of the largest leaf was decreased in the BpXINl-120, pMPSl3-7, and
pXINl-ZSB transgenic lines compared to wild type during the first replication (Table

4.1). The longest leaf on wild type plants measured 17.7 cm from node to leaf tip, while

92

BpXIN1-120, pMPS]3-7, and pXIN1-2SB had leaf lengths of 15.9, 10.9, and 15.9 cm,
respectively. Leaf width was also altered in several lines (Table 4.1). pXIN1-17B leaves
were wider than wild type, measuring nearly 6.5 cm across, compared to 5.5 cm for wild
type. The largest leaves of pMPS]3-7 and pXIN 1-258 were significantly narrower than
wild type, measuring 2.4 and 5.0 cm, respectively. In the second replication, pMPSl3-7
again had shorter and narrower leaves compared to wild type. pMPS]3-10 had slightly
wider leaves and pXIN 1-258 had slightly narrower leaves than wild type in this
replication (Table 4.2).

Flower size was reduced in the pMPS 1 3-7 line in both replications (Tables 4.1
and 4.2). In the first replication, length of the first fully open wild type flower was 5.4
cm from corolla rim to base of the receptacle, while pMPS 1 3-7 flowers were 4.9 cm.
Corolla rim diameter was also reduced from 5.4 cm in wild type to 4.3 cm in pMPSl3-7.
'In the second replication, the first fully open pMPSl3-7 flowers measured 4.5 cm long
with a corolla diameter of 4.1 cm. This is significantly smaller than wild type flowers in
this replication which were 5.5 cm long with a corolla diameter of 5.2 cm.

Aboveground biomass varied by genotype and replication (Tables 4.1 and 4.2).
In the first replication, biomass of pMPS]3-10, BpXINl-l 10, pMPS]3-7, and pXIN1-
253 was 9.8, 10.4, 6.4, and 7.5 g, respectively, compared to 12.9 g for wild type.
However, the empty vector control plants also had reduced biomass compared to wild
type with an average mass of 9.5 g. In the second replication, only pMPS]3-7 differed

from wild type, with a biomass of 11.6 g compared with 8.7 g for wild type.

93

DISCUSSION AND CONCLUSIONS

Heterologous expression of CBF transcription factors results in increased freezing
tolerance of many species (Jaglo et al. 2001, Kasuga et a1. 1999, Owens et al. 2002, Pino
et al. 2007, Pino et al. 2008). However, the use of the strong constitutive CaMV 35S
promoter to drive expression of CBF often causes undesirable effects such as stunting
and delayed flowering (Liu et a1. 1998; Kasuga et al. 1999; Gilmour et al. 2000; Zhang et
a1 2004, Pino et al. 2007). In tomato, constitutive expression of AtCBF 1 reduces fruit
number, seed production, and plant biomass (Hsieh et al. 2002). In potato, constitutive
expression of AtCBF 1 -3 reduces biomass and tuber yield (Pino et al. 2007). Here we
have shown that improved freezing tolerance of petunia through constitutive expression
of AtCBF 3 comes at a cost to horticultural traits, especially flowering time.

Long periods between planting and flowering are undesirable in greenhouse crops
due to the increased time and money that a grower must invest. High expression of
AtCBF 3 in the pMPS 1 3-7 transgenic line delayed flowering by several weeks, increasing
time to flowering by more than 30% in the first and 65% in the second replications
compared to wild type plants. Expression of AtCBF 3 or LeCBF] in the other transgenic
lines did not delay flowering, but none of these lines showed increased freezing tolerance
(chapter 3).

The length of the flower-bearing stem was reduced by more than half in the
pMPS]3-7 line in the first replication, with an average length of 15.6 cm compared to
32.2 cm for wild type. While the flower-bearing stem was shortened, the number of
leaves present on the stem was unchanged, indicating that reduced stem length was the

result of shortened intemodes. Although the pMPS]3-7 plants flowered several weeks

94

tr-—

 

after wild type, the leaf/node numbers of both the flower-bearing and primary stems
reveal that flowering still occurred at the same physiological point, but reduced
development rate resulted in delayed flowering. Several other transgenic lines had
reduced stem length, but none as extreme as pMPSl3-7. The number of leaves below the
first flower on these stems was also unchanged relative to wild type.

The number of visible buds was dramatically reduced in pMPS 1 3-7 in the first
replication, an undesirable result in petunia, a plant grown primarily for its abundant
flowers. In the second replication, pMPS]3-7 had slightly more visible flower buds at
time of flowering, but this is likely a result of the additional 50 days of growing time that
this line had relative to wild type. Flower length and width of the pMPS]3-7 line was
also reduced in both replications.

In addition to changes in flowering characteristics, the pMPS 1 3—7 line was
smaller in overall size and displayed a more prostrate growth habit with lateral branches
that grow parallel to the ground rather than curving upward (Figure 4.1). The primary
stem of this line was also very short, with an average length of just 3.3 cm at time of
flowering. Several transgenic lines showed altered leaf length and width. Most notably,
the largest leaves of pMPS]3-7 were significantly shorter and narrOwer than wild type in
both replications.

Aboveground dry biomass measurements were highly variable between different
transgenic lines and also between replications of the same line. In the first replication,
pMPS]3-7 had the lowest biomass at the time of flowering, only half as much as wild
type, while changes in biomass of other lines were not as dramatic. In the second

replication, pMPS 1 3-7 had increased biomass compared with wild type.

95

Heterologous expression of AtCBF 3 in P. hybrida can increase freezing tolerance
if expressed at high levels (chapter 3). Unfortunately, plants expressing the transgene at
these levels are also impacted by negative phenotypic changes. The pMPS]3-7 line is
more tolerant to freezing temperatures than wild type, but this line would not be
marketable as a horticultural crop. Further work should be done to examine the
feasibility of using a stress-inducible promoter to drive expression of AtCBF 3. This
technique has successfully been used to minimize the negative effects of constitutive
CBF expression, while maintaining the benefits of increased freezing tolerance (Kasuga

et a1. 1999, Pino et al. 2007).

96

55.20:: 30:55 3:805 0: @6300: 5035095: . _o. :5 morn: .0 0%: 23 E0: 805:6 >_::00:_:w_m 0:00:05 5.. :5 ..

 

 

5mm 3.th WNN ~.mm fix :6 Wm we Ohm m._ m0: ONT—253m
w.mm 5m. _ m film *odm 0.” N6 Wm «6 NR M: «.3. 2 7.253:
mém Imdm 0.3 dem :5 06 0m *0: 0.0m v; wdw mmm-_Z_X:
Yam 30.9“ Sam :3. 2: Nd 9m 06 :N .0.— n._w mun—42:3
Iodm w._~ LEN 56$ 0.: 9m 9m 0.: 04mm _._ 0:. STm—mmEQm
.1.ch *ewd— m.w— **w.w_ Nu. _.m Wm {*Nd QWN va— **v.h© oTM—mL—ZQ
9mm _.:..m.m 5mm vém .33.: 3L: 1%.? **M.N 1.0.x *m._m v; _.:..N.N.N_ brawn—)3
wém Ind. 0. _m Ndm 0.: Wm Wm :0 mdm m._ v.2. ._050> 335
Wow v.3 m.- 0.8 5.x m6 m6 mé vém m..— wg. 0%. 25’
50:5: AEov 59:5: AEov @ A83 A68 Afiuv A53 50:5: 5:80: 530: e “:05
:00. E05. 5w:0_ :00. E05. 5w:0_ 30.: 55:56 fiwfi: £23 5w:0_ :0: .530: 0: 930
@255 E03 .5307. E03 -05 0:800 530: :00.— .304 530E :0:0
.008?— 530E LOO .530: at... 5:0 Er. .

 

.:0m0 DE: 33 530: 5:: 0.: .3: 05
:85: 0.53 358055.008 =< .:0::0=a0: “0:805. :_ .=0:8_2. 033% 0.350% 090 “0:3 :5 0_:0wm:0.: :0 3:0: 35:03.5: :0 Swish—E00 .Né 03:...

55:20:: 30:55 “5:803: 0: @8800: 5035013: .5. :5 3.00 :0 0:5 “0:3 :5: 605:6 3:005:30 0:00:05 1. 0:: ..

 

 

_.mm ofim ma: .6 Wm Wm 1.0.2 mSN N..— fimn ONT—233m
a. _N 0.3 ll: ~.m Wm fin ms. vdm 0; ed: 3 2253:
vdm 5.3 .22. Nm m.m *o.m 3.0.3 0.8 v; as: mm~-_Z_Xq
Ymm 0.0m 0.2 9m 9m 1.0.0 0.: YNN o._ :3. m: 7.758
:mm 0.0m N: :6 ~.m :6 QE :3. 0.. ad: 8722315
wdm m.mm Ind m.m m.m m.m W: Dam 0.. 0.21 on—mmv/E
0.2 1.0.2 1.0.0 1%.: 3.: 33% 2.0.3 1...: 0; 3.5.3 WET—SE
a. 5 Wow 1.00 :6 m.m 06 00— mdm .1 ad: :050> .935
0.2.. m.~m Num— vd :6 9m Nu: m.wm 0N 0.2 090 2:5
5:83: 2:3 va AES AEov A63 A68 50:5: 59:0: .530: 9 E03
:00. E05 .35: $9: .5585: 5w:0_ 523 5w:0_ :0: 530: 9 £00
530E :55 -05 0:88 530: :00.— .304 .530."— :25
.532..— bo .530: 5:: :0:0 5:...

 

.:0m0 3:: 5.03 .530: 5:: 0.: .8: 05 :83 0.53 3:0:550006
:< dogma—:0: 5:: :_ 20:022. 0333.: 3250:: 093 :23 :23 8:: 3:03:00 :0 £5: _::==:0E0: :0 5:59:00 A... 030:.

97

.0505 0::E:x0:::: 08:5 0:: 0: :0N5 0:: 50:08: 5:»: 2:3 :05: 55:5:00 :03
:0: :0550::x0 :mU 30:20:05: :0 _0>0_ 50::wE 0:: 5:3 50:: 0::0wm:::: 0:: ::0:: 51:30:33: 0>:::::0m0::0: 0:: :30:m
.:0:0:0 :030: 5:: 0:: .3: 0:: :0 0:»: 2:3 95 50:: 35:55:: :0 0:50:05: w:::::::00 m::::w0:0:: 4.: 03:3

 

 

0:3. 25> A :lm004....mmmv m n T w 2.x: Amlm0:<....mnmv Wm 5:2:

   

 

 

98

REFERENCES

Gilmour S., D. Zarka, E. Stockinger, M. Salazar, J. Houghton, and M. Thomashow. 1998.
Low temperature regulation of the Arabidopsis CBF family of AP2 transcriptional
activators as an early step in cold-induced COR gene expression. Plant J. 16:433-
442.

Gilmour S., A. Sebolt, M. Salazar, J. Everard, and M. Thomashow. 2000. Overexpression
of the Arabidopsis CBF 3 transcriptional activator mimics multiple biochemical
changes associated with cold acclimation. Plant Physiol. 124:1854-1865.

Gilmour S., S. Fowler, and M. Thomashow. 2004. Arabidopsis transcriptional activators
CBF 1, CBF2, and CBF 3 have matching functional activities. Plant Mol. Biol.
54:767-781.

Guy C., K. Niemi, R. Brambl. 1985. Altered gene expression during cold acclimation of
spinach. Proc. Natl. Acad. Sci. 82:3673-3677.

Hsieh T., J. Lee, Y. Charng, and M. Chan. 2002. Tomato plants ectopically expressing
Arabidopsis CBF] show enhanced resistance to water deficit stress. Plant Physiol.
130:618-626.

Jaglo K., S. Kleff, K. Amundsen, X. Zhang, V. Haake, J. Zhang, T. Deits, and M.
Thomashow. 2001. Components of the Arabidopsis C-repeat/dehydration-
responsive element binding factor cold-response pathway are conserved in '
Brassica napus and other plant species. Plant Physiol. 127:910-917.

J aglo-Ottosen K., S. Gilmour, D. Zarka, O. Schabenberger, and M. Thomashow. 1998.
Arabidopsis CBF] overexpression induces COR genes and enhances freezing
tolerance. Science 280: 104-106.

Kasuga M., Q. Liu, S. Miura, K. Yamaguchi-Shinozaki, and K. Shinozaki. 1999.
Improving plant drought, salt, and freezing tolerance by gene transfer of a single
stress-inducible transcription factor. Nature Biotechnol. 17:287-291.

Liu Q., M. Kasuga, Y. Sakuma , H. Abe, S. Miura, K. Yamaguchi-Shinozaki, and K.
Shinozaki. 1998. Two transcription factors, DREBl and DREB2, with an
EREBP/APZ DNA binding domain separate two cellular signal transduction
pathways in drought- and low-temperature-responsive gene expression,
respectively, in Arabidopsis. Plant Cell 10: 1 391-1406.

Owens C., M. Thomashow, J. Hancock, and A. Iezzoni. 2002. CBF] orthologs in sour

cherry and strawberry and the heterologous expression of CBF] in strawberry. J.
Amer. Soc. Hort. Sci. 127:489-494.

99

Pennycooke J ., M. Jones, and C. Stushnoff. 2003. Down-regulating a-Galactosidase
enhances freezing tolerance in transgenic petunia. Plant Physiol. 133:901-909.

Pino M.T., J .S. Skinner, E.J. Park, Z. Jeknic, P.M. Hayes, M.F. Thomashow, and T.H.H.
Chen. 2007. Use of a stress inducible promoter to drive ectopic AtCBF expression

improves potato freezing tolerance while minimizing negative effects on tuber
yield. Plant Biotechnol. J. 5:591-604.

Pino M.T., J .S. Skinner, Z. Jeknic, P.M. Hayes, A.H. Soeldner, M.F. Thomashow and
T.H.H. Chen . 2008. Ectopic AtCBF] over-expression enhances freezing tolerance

and induces cold acclimation-associated physiological modifications in potato.
Plant Cell Environ. 31 :393-406.

Stockinger E., S. Gilmour, and M. Thomashow. 1997. Arabidopsis thaliana CBF]
encodes an AP2 domain-containing transcriptional activator that binds to the C-
repeat/DRE, a cis-acting DNA regulatory element that stimulates transcription in

response to low temperature and water deficit. Proc. Natl. Acad. Sci. 94:1035-
1 040.

Thomashow M. 1999. Plant cold acclimation: Freezing tolerance genes and regulatory
mechanisms. Annu. Rev. Plant Physiol. Plant Mol. Biol. 50:571-599.

Yelenosky G., and C. Guy. 1989. Freezing tolerance of citrus, spinach, and petunia leaf
tissue. Plant Physiol. 89:444-451.

Zhang X., S. Fowler, H. Cheng, Y. Lou, S. Rhee, E. Stockinger, and M. Thomashow.
2004. Freezing-sensitive tomato has a functional CBF cold response pathway, but
a CBF regulon that differs from that of freezing-tolerant Arabidopsis. Plant J.

39:905-919.

100

CHAPTER 5

ASSESSING DROUGHT TOLERANCE OF CBF OVER-EXPRESSING PE T UNIA
H YBRIDA ‘MITCHELL’

ABSTRACT

Ectopic expression of CBF transcription factors has been reported to improve
drought tolerance in several species, including Solanum lycopersicum (tomato). The
objective of this work was to determine whether constitutive heterologous expression of
AtCBF 3 from Arabidopsis thaliana or LeCBF] from S. lycopersicum in Petunia hybrida
results in increased tolerance to drought stress. Following 9 d of water withholding,
survival of the transgenic lines was not different than wild type. After 12 d of water
withholding, survival of the transgenic line with the highest expression of AtCBF 3,
pMPS]3-7, was lower than wild type, while the survival of all other lines was not
different. No differences in relative biomass gain were seen between wild type and

transgenic lines following either 9 or 12 d of water withholding.

INTRODUCTION
Water deficits are a major stress facing plants in the garden. Recent water
shortages, especially in the southeastern and western parts of the country, highlight the
need for ornamental plants that are able to thrive under low water conditions. An
increasing population (US Census Bureau 2004), coupled with decreasing water

resources, will likely lead to increased water use restrictions in this country. It is

101

therefore wise to develop ornamental plants that can tolerate or thrive in low water
conditions.

Drought tolerance is a complex trait involving many genetic factors (Shinozaki et
al. 2003). Among these is the CBF family of transcription factors isolated from
Arabidopsis thaliana L. consisting of three cold-induced members, AtCBF1-3 (Gilmour
et al. 1998), and a drought-induced member, AtCBF 4 (Haake et al. 2002).
Overexpression of AtCBF 3 (Liu et al. 1998; Kasuga et al. 1999) or AtCBF 4 (Haake et al.
2002) in Arabidopsis increases both drought and cold tolerance of transgenic plants.

Many species encode homologs of the Arabidopsis CBF genes, including
Solanum lycopersicum L. (Zhang et al. 2004), Brassica napus L. (Jaglo et al. 2001; Gao
et al. 2002), Prunus avium L. (Kitashiba et al 2004), and Zea mays L. (Qin et al 2004).
Heterologous expression of AtCBF 1 in S. lycopersicum L. increases drought tolerance
(Hsieh et al. 2002) and expression of AtCBF 1 , 2, or 3 enhances drought tolerance of
Oryza sativa L. (Ito et al. 2006).

The objective of this study was to determine whether constitutive heterologous
expression of AtCBF 3 or LeCBF] (a CBF homolog from S. lycopersicum) enhances the

drought tolerance of transgenic Petunia hybrida.

MATERIALS AND METHODS
Analysis of drought tolerance of transgenic lines
Transgenic P. hybrida ‘Mitchell’ lines over-expressing AtCBF 3 or LeCBF]
behind the constitutive cauliflower mosaic virus (CaMV) 3 5S promoter were created as

previously described in chapter 3. These lines were tested to determine the impact of

102

heterologous CBF expression on drought tolerance. Seeds were surface sterilized by
soaking for 10 to 15 min in a 50% bleach, 0.1% Triton-X-IOO solution with gentle
agitation. Seeds were rinsed with sterile deionized water 3 to 4 times to remove all traces
of bleach and then suspended in a 0.1% sterile agar solution to facilitate pipetting. Seeds
from transgenic lines were pipetted onto 100 x 15 mm disposable Petri plates containing
Gamborg’s B5 medium (minus sucrose, kinetin and 2,4-D; plus 0.7% agar) supplemented
with 100 pg/ml kanamycin. Seeds from wild-type control plants were pipetted onto
identical plates without the addition of kanamycin. Seeds were plated at a density of
~100 seeds per plate. Plates were placed in a 22 °C, 16 h photoperiod (100-130 pmol rn'2
3’1 light) chamber for three weeks. After three weeks, three plants of each genotype
including wild type and an empty vector control were transplanted in each of twenty 27 x
54 cm open flats filled with 4 cm of 70% peat moss, 21% perlite, 9% vermiculite
(Suremix, Michigan Grower Products Inc., Galesburg MI, USA). Plants were placed
randomly in each flat with a spacing of 5 cm between each plant. After one additional
week in a 22 °C, 16 h photoperiod growth chamber with increasing light intensity to
harden the plants off, the trays were transferred to a greenhouse at 20 °C with a 16 h
photoperiod for 10 d in order to allow for the establishment of the plants. At the onset of
the experiment, the plants from four trays were harvested for aboveground biomass
measurement to determine relative initial plant size. The soil in the remaining trays was
saturated to ensure uniform initial soil moisture conditions. Water was then withheld
from four trays for each drought period, 9 d or 12 d, after which the soil was again
saturated and plants were allowed to recover for one week. At the conclusion of the

recovery period, survival and final biomass were determined. During the drought period,

103

four trays were watered normally to be harvested as unstressed controls along with the
drought stressed plants for each time period. This experiment was conducted twice.
Fisher’s Exact Test was conducted to determine whether survival of transgenic lines
differed from wild type. ANOVA analysis of the relative biomass gain for each line was

conducted using PROC GLM with SAS (SAS Institute Inc., USA) software.

RESULTS

Heterologous expression of CBF does not confer drought tolerance

Survival of P. hybrida under drought conditions was not enhanced by high
heterologous expression of CBF transcription factors. In fact, under these experimental
conditions, the lowest survival rates were seen in lines expressing the transgenes at the
highest levels (Table 5.1). Though not statistically different than wild type, the only
plants that did not survive the 9 day drought in the first replication were pMPS 1 3-7,
while all plants survived the second replication. After 12 days of drought, the only plants
that did not recover in the first replication were from the pMPS]3-7 and pXINl-17B
lines, although only the survival of pMPS 1 3-7 was statistically less than wild type.
Likewise, in the second replication pMPS]3-7 and pXIN1-17B had the lowest survival,
but only pMPS]3-7 was significantly lower than wild type. It should be noted that
pMPS]3-7 appears to have a much less extensive root system than wild type plants which
may have contributed to the reduced survival under low water conditions.

In addition to survival data, relative biomass gain was used to determine how well
the transgenic lines grew under water stress. Direct comparisons of biomass before and

after drought are difficult because of the severe stunting seen in the pMPS 137 line. The

104

pMPS]3-7 line was smaller than wild type at the onset of the drought period and even
under ideal conditions grew slower. Therefore, comparisons were made based on
“relative gain” which was calculated as the percent of unstressed biomass gain that was
achieved by stressed plants during the drought and recovery periods (Table 5.2). There
was no significant variation between the biomass gained by the transgenic lines and the

controls following either of the drought periods (Tables 5.3 and 5.4).

DISCUSSION AND CONCLUSIONS

Heterologous expression of CBF did not improve the survival of P. hybrida in our
water withholding experiments. Survival of pMPSl3-7 was significantly less than wild
type following 12 days of water withholding. This may be due to the less extensive root
system that was present on this high expressing transgenic line. The pXIN1-17B line
also seems to have a less aggressive root system, but the survival of this line was not
statistically different from wild type. This was a competitive survival study where
transgenic lines were grown alongside wild type plants. The pMPSl3-7 line is a severely
stunted line that grows very slowly. This may have also reduced its survival because it
was shaded by the taller wild type plants. However, this is not true for pXIN1-17B which
has normal wild type growth. Under drought conditions typically encountered by a plant
in the garden, it seems that the negative impact of high CBF expression on root growth
may outweigh any potential benefit on drought tolerance at the cellular level due to CBF
expression.

There was significant biomass variation between the two replications of this

experiment, hindering the detection of differences between transgenic lines and controls.

105

There was no significant difference between the relative gain values for any of the lines
(Tables 5.3 and 5.4). In the first replication, it appears that the pMPS]3-7 line was less
tolerant of water stress than wild type during 9 days of drought (73 % relative gain for
pMPS]3-7 vs. 93 % for wild type), but in the second replication it appeared to be more
tolerant than wild type (336 % relative gain for pMPS]3-7 vs. 109 % for wild type)
(Table 5.2). The same was true for the 12 day drought period where wild type had 21 and
45 % relative gain in the first and second replications, respectively, while pMPS]3-7 had

15 and 108 % (Table 5.2).

106

Table 5.1. Survival rates of transgenic P. hybrida 'Mitchell' plants
expressing AtCBF 3 (pMPSl3 and BpMPSl3) or LeCBF] (pXINl and
BpXINl) following 9 or 12 days of water withholding and 7 days of
recovery. Numbers shown represent percent survival of 12 individuals.

 

 

 

 

9 day drought 12 day drought

Genotype Rep 1 Rep 2 Rep 1 Rep 2
Wild type 100 100 100 92
Empty vector 100 100 100 100
pMPSl3-7-12 83 100 25* 17*
pMPSl3-10-8 100 100 100 92
BpMPS13-101 100 100 100 100
pXIN1-17B-20 100 100 75 58
pXIN1-25B-12 100 100 100 83
BpXIN1-110-5 100 100 100 83
BpXINl -120-2 100 100 100 92

 

* indicates significant variation from wild type survival as tested by
Fisher's Exact Test (alpha = 0.05)

Table 5.2. Relative gain of above-ground dry biomass averaged for 12 plants per
genotype in each drought treatment. Relative gain of drought plants was defined
as the percent of non-stressed weight gain achieved by stressed plants; calculated
as: Relative gain = (droughtfinal - initial) / (nonstressedfinal - initial) X 100

 

Relative gain (%)

 

 

 

9-day drought 12-day drought
rep 1 rep 2 rep 1 rep 2
Wild type 93 109 21 45
Empty vector 59 75 38 44
pMPSl3-7-12 73 336 15 108
pMPSl3-10-8 79 148 28 45
BpMPSl3-101 87 66 39 58
pXIN1-17B-20 158 72 15 36
pXIN1-25B-12 106 121 58 69
BpXINl-110-5 83 81 32 54
BpXIN1-120-2 90 58 ' 33 51

 

107

Table 5.3. ANOVA for effect of transgenic line on relative biomass gain for 9-day

drought period.

Dependent variable: Sum of Mean

Relative Gain (%) Source DF Squares Square F Value Pr > F
Model 8 27813.11 3476.64 0.75 0.65
Error 9 41766.00 4640.67
Total 17 69579.1 1

Table 5.4. ANOVA for effect of transgenic line on relative biomass gain for 12-day

drought period.

Dependent variable: Sum of Mean

Relative Gain (%) Source DF Squares Square F Value Pr > F
Model 8 2480.00 310.00 0.49 0.83
Error 9 5640.50 626.72
Total 17 8120.50

108

REFERENCES

Gao M., G. Allard, L. Byass, A. Flanagan, and J. Singh. 2002. Regulation and
characterization of four CBF transcription factors from Brassica napus. Plant
Mol. Biol. 49:459-471.

Gilmour S., D. Zarka, E. Stockinger, M. Salazar, J. Houghton, and M. Thomashow. 1998.
Low temperature regulation of the Arabidopsis CBF family of AP2 transcriptional

activators as an early step in cold-induced COR gene expression. Plant J. 16:433-
442.

Haake V., D. Cook, J. Riechmann, O. Pineda, M. Thomashow, and J. Zhang. 2002.
Transcription factor CBF4 is a regulator of drought adaptation in Arabidopsis.
Plant Physiol. 130:639-648.

Hsieh T., J. Lee, Y. Charng, and M. Chan. 2002. Tomato plants ectopically expressing
Arabidopsis CBF] show enhanced resistance to water deficit stress. Plant Physiol.
130:618-626.

Ito Y., K. Katsura, K. Maruyama, T. Taji, M. Kobayashi, M. Seki, K. Shinozaki, and K.
Yamaguchi-Shinozaki. 2006. Functional analysis of rice DREBI/CBF-type

transcription factors involved in cold-responsive gene expression in transgenic
rice. Plant Cell Physiol. 47:141-153.

Jaglo K., S. Kleff, K. Amundsen, X. Zhang, V. Haake, J. Zhang, T. Deits, and M.
Thomashow. 2001. Components of the Arabidopsis C-repeat/dehydration-
responsive element binding factor cold-response pathway are conserved in
Brassica napus and other plant species. Plant Physiol. 127:910-917.

Kasuga M., Q. Liu, S. Miura, K. Yamaguchi-Shinozaki, and K. Shinozaki. 1999.
Improving plant drought, salt, and freezing tolerance by gene transfer of a single
stress-inducible transcription factor. Nature Biotechnol. 17:287-291.

Kitashiba H., T. Ishizaka, K. Isuzugawa, K. Nishimura, and T. Suzuki. 2004. Expression
of a sweet cherry DREBI/CBF ortholog in Arabidopsis confers salt and freezing
tolerance. J. Plant Physiol. 161:1171-1176.

Liu Q., M. Kasuga, Y. Sakuma, H. Abe, S. Miura, K. Yamaguchi-Shinozaki, and K.
Shinozaki. 1998. Two transcription factors, DREBl and DREB2, with an
EREBP/AP2 DNA binding domain separate two cellular signal transduction

pathways in drought- and low-temperature-responsive gene expression,
respectively, in Arabidopsis. Plant Cell 10:1391-1406.

Qin F ., Y. Sakuma, J. Li, Q. Liu, Y. Li, K. Shinozaki, and K. Yamaguchi-Shinozaki.
2004. Cloning and functional analysis of a novel DREBl/CBF transcription factor

involved in cold-responsive gene expression in Zea mays L. Plant Cell Physiol.
45:1042-1052.

109

Shinozaki K., K. Yamaguchi-Shinozaki, and M. Seki. 2003. Regulatory network of gene
expression in the drought and cold stress responses. Current Opinion Plant Biol.
6:410-417

US Census Bureau. 2004. US. Interim Projections by Age, Sex, Race, and
Hispanic Origin. <http://www.census.gov/ipc/www/usinterimproj/>

Zhang X., S. Fowler, H. Cheng, Y. Lou, S. Rhee, E. Stockinger, and M. Thomashow.
2004. Freezing-sensitive tomato has a functional CBF cold response pathway, but

a CBF regulon that differs from that of freezing-tolerant Arabidopsis. Plant J.
39:905-919.

110

MICHI AN STATE UNIVERSITY LIBRARIES

III III IIIHI |

3 129313062 6471